It’s been a rough couple of weeks for the Space Kadet program. But after launching several thousand tonnes of ships — and many, many lessons learned — Space Station “Origin” is finally complete.
This is a long one, so buckle up and read on to watch me fumble my way to victory. Eventually.
Building a Space Station
Armed with only a very basic understanding of the MKS/OKS life support loop from their wiki, I designed this simple station core. In the middle are the three modules needed for a basic life support loop:
- The Kerbitat turns waste into compost, and turns biomass into food.
- The Aeroponics Bay uses compost to grow plants and produce biomass. It also uses these plants to purify water and convert carbon dioxide back into oxygen.
- The Colony Command Centre is required for the others to work, and its control capability is represented by producing a “punchcards” resource that the other modules consume.
- The Science Bay isn’t part of the life support loop, but I needed a fourth module, and it allows us to process science data on-site for a 5x boost to its transmission value.
The habitat rings and agricultural modules (greenhouses) would sit symmetrically around these core components.
I wanted to have rotating habitat rings for artificial gravity, even if it was just for show and wouldn’t have any in-game effect. So I added some Infernal Robotics “rotating washers” to manually rotate the habitat rings. The idea was that they would rotate in opposite directions, cancelling each other out and not twisting the station too much.
(Of course, what I didn’t realise at the time was that OKS habitat rings would slowly rotate on their own, making the washers a bit useless. And since I wasn’t prepared for that, they ended up rotating both in the same direction. D’oh! Not that it matters, but I prefer real physics where possible.) 😦
Now, there was no way I was going to get something this heavy and weirdly-shaped into orbit all in one piece. Since the aeroponics wings formed the longer of the two axes, I broke those off and transported it in three separate trips. (You can see the docking ports where I plan to reattach them in the diagram above.)
But of course, like any new heavy construction project, the hard part is getting it into orbit.
Most of the Sky Krane series of ships looked something like this. As per traditional rocket design, they started out narrow at the top, and then flared out for a wider base, this time using a quad-coupler and four engine stacks.
To deposit the payload (inside the fairings), we would
- undock the command module and payload from the engines,
- dock the payload to the station,
- undock the command module from the payload,
- dock the command module to the engines and return home.
We ended up launching a total of 19 Sky Kranes, and almost every one taught us something new:
Sky Krane 2 pointed out an interesting side effect of fairings — if anchored at both the top and bottom (as in the diagram above), they can provide significant structural stability. When they auto-blew at 75km, the ship turned into a giant Space Slinky — it was bending nearly 180º and almost hitting itself! Needless to say, we did a sub-orbital abort.
Sky Krane 5 taught us that you can do a 360º backflip over the launchpad and still successfully make it into orbit. We deploy the middle section, including the two hab rings.
Sky Kranes 6 and 7 demonstrated that aerodynamic issues can actually get worse as your ship gets lighter. At the minimum, your ship must have at least one of high density, good aerodynamics, or a low thrust-to-weight ratio. Trying to power your way through the atmosphere with a broad, light ship results in mayhem.
Sky Krane 8’s long and costly correction burn is a reminder that we should’ve checked our inclination before dropping off the first base component (it was over 6º) — and, failing that, we should time and orient our launches to match the inclination, instead of launching in an equatorial orbit and correcting it later. We dock our first wing to the station.
Sky Krane 10 reminded us how dangerous high-speed atmospheric aborts can be. We lost our heat shield and one of the command pod’s solar panels, but everyone survived.
Sky Krane 11 again showed off with a 360º backflip, but it also demonstrated the pitfall of waiting until we reach the station to discard our stages and fairings. There’s now a huge debris field on the same orbit as Origin Station, and due to a mis-click, we ended up rendezvousing with a discarded fairing rather than the station. Thankfully, we had the fuel (and life support time) to readjust and meet the station.
Sky Krane 12 found an interesting use for the Mark 2 Lander Can — if you attach a large docking port to the large flat top, it can be very useful as a manned heavy station pusher. We used it to correct the station’s inclination burn, getting it to a perfect equatorial orbit.
Sky Kranes 13 and 14 finally show us that you can only push an unstable ship design so far before it’s time for a rethink. We’re trying to lift a solar panel array, but it’s far too light, and the ship doesn’t have enough inertia to overcome the aerodynamic forces.
Time for a rethink.
Sky Krane 15 demonstrates the old adage, “the solution to any KSP problem is to add more struts.” Multiple stacks heavily strutted together are a wonderfully stable solution, even if it’ll be bit tricky to swap out the payload.
Mega Lifter 1 shows off the value of the Hangar mod’s “Multipurpose Resource Tank”, basically a giant tank that you can edit and subdivide at will. Weighing in at a total of 2147 tonnes, it allows us to get 117 tonnes of “RocketParts” to the station — enough to build 117 tonnes of ships — but due to the reconfigurability of the tank, it can potentially serve as a template for getting many different kinds of resources into orbit.
Sky Krane 17 brings the orbital shipyard to the station. Given how difficult it would be to abort or eject, it also marks our first use of the Deployable Emergency Reentry Pod (DERP) — basically a lifeboat with its own propulsion and parachute, and easy evacuation via a hotkey. The ship is stable and we never need it, though, so I still don’t have any hands-on experience with it.
Sky Krane 18 demonstrates that our old Sky Krane design could be quite stable if you have only a few heavy components (and add some struts just in case). It also redefines “dense,” as that one little barrel contains 16 tonnes of machinery and 16 tonnes of spare parts for the life support systems — no wonder they’re empty by default!
We’re out of docking ports, so I design this ship to be docked at both ends, sitting inbetween the station and an existing docked ship. Those extra 1.25m docking ports are in case we need to dock anything else, and they pay off when we later need to add some temporary storage for the compost and biomass. (Even temporary intermediate products need a place to stay.)
H-H-H-Houston We Have a P-P-P-Problem
With our life support system running and our shipyard operational, we’re finally ready to start building some larger station wings, right in place at the station itself.
Our first extension is the Docking Wing, replacing the existing shipyard. Right now, everything has to be self-propelled because we don’t have the technology to make miniature “tugs” to carry things around, so two of the docking ports are carrying temporary monopropellant tanks.
I forget to put generators or batteries on it, so it quickly runs out of power and starts drifting away. After a collision and a dicey recovery, we finally manage to get the old shipyard docked to it, supplying it with power and letting us dock it with the main station.
The trouble doesn’t end there, though. Once in place, removing the monopropellant tanks and the old shipyard causes the Docking Wing to start oscillating and flailing around, not unlike one of those fan-inflated “waver” dolls you find outside as advertisements.
That’s fun to watch when it’s colourful nylon, but not so much when it’s hundreds of tonnes of metal! The whole station is shaking, and left unchecked, we risk killing the crew with g-forces.
Thankfully, we manage to bolt everything down by sending Jeb out with a box of Quantum Struts and applying them liberally across the entire structure. They’re basically energy-based “tractor beams” that can be turned on and off at will, and once installed, they provide the extra strength and stability we need.
Piece By Piece
With our new efficient high-capacity shipyard, we design and build the rest of the space station:
As my attempt at a combined storage and power solution, the Storage Wing turns out to be a bit of a failure. We dock it to one of the habitat rings, thinking that the hab ring’s rotating washer can rotate it to face the sun along one axis, while its own rotating washers can rotate on the other axis.
Turns out, rotating many tonnes of storage is a very bad idea, especially when they’re full and heavy. The entire station torques every time we do, they take up to a minute to settle out after each rotation, and the hab ring we’re attached to is starting to glitch and threatening to rip off the station after several adjustments.
The new Science Wing attempts to offload storage from the Storage Wing so that we can replace it. But more importantly, it also contains the labs and cyclotrons (particle accelerators) we’ll need to do orbital experiments. At the tip, I add an emergency lifeboat with 18 passenger slots and a nuclear reactor.
Having learned our lesson from the Docking Wing, there are a ton of Quantum Struts holding the whole thing rigid, and it ends up being quite stable despite the extreme mass. (The zoology bay is 20 tonnes, and each of the two cyclotrons is 30!)
With the resources offloaded, we now seek to replace the Storage Wing with the Omnidirectional Solar Array, something I hope will provide enough power no matter what direction the sun is in, without rotating. (Life support systems turn out to be an incredible power drain!)
Unfortunately, it’s a bust — the panels are great in certain orientations, but rapidly turn crappy again as they obscure each other. Plus the part count is through the roof, adding tons of lag to an already laggy station.
Instead, I try turning on the nuclear reactor, and I discover that it produces a ridiculous amount of power and will last us for years. We’re already having trouble with our high part count, so I decide to ditch the solar — we use its monopropellant to put it in a lower orbit, where it will become a (ridiculously overengineered) communications satellite instead.
With our power crisis solved and the labs now in operation, we’re finally able to begin doing some orbital science.
All of the experiments involve some combination of Eurekas (from scientists in the Science Lab), Kuarks (from the Cyclotrons), and Bio-Products (from the Zoology Bay, and almost certainly a euphemism for poop). In general, you start the experiment, accumulate the correct number of these — Kuarks decay, so you need a certain minimum production — and then finalise the results, where the experiment becomes unusable and you get a standard science report.
The standard Station Science experiments normally only give you 20% for transmitting, but our Science Bay lets us process these for 5x the value, so we’re able to transmit these home and then discard them to a lower orbit. The Fuel Science experiments are only 10% value and would only net us 50% if processed locally, so we send those back to Kerbin with a parachute and pick them up.
Meanwhile, maintaining the animals in the Zoology Bay is providing a slow trickle of Science — about 1 Science per day, but with diminishing returns over time.
So what technologies do we buy with all this Science?
With the 300 Science from the first batch of station experiments, we grab Nuclear Propulsion. Nuclear engines are the most efficient moderate-thrust engines available, but they’re heavy and terrible in the atmosphere. Since we’re building in space, most of their disadvantages are gone now, and they’ll likely become our primary means of getting large ships around.
Our second batch of experiments focuses on fuels, and we get about 550 Science total. That’s enough to pick up Precision Engineering (160) and Ion Propulsion (300), so we can start sending efficient interplanetary probes out now.
Getting ion also unlocks a xenon-based fuel experiment, which gets us another 270 Science. With a few extra points we have lying around, I grab Field Science (300), which focuses on rovers and will give us something to do on the Mun and Minmus while our probes are in transit.
So in the end, we have our completed space station … but I’m not particularly happy with it.
For one, the lag is awful. That’s partly because the base is big and does a lot of things at once, but also partly because it’s assembled in space and is a bit haphazard, not to mention all the inter-wing docking ports. I know there’s some inherent KSP lag just from having ships or stations with large part counts, and my extensive mod list probably just makes that worse.
With the technology now at hand, we have the ability to use the Logistics Hub, a drone-based delivery system for automatically shuttling resources around inside the same “Sphere of Influence” (e.g. within Kerbin orbit), including from the surface to orbit. That makes resupplying the base incredibly easy, which means we can cut out the entire life support system and save a lot of mass and complexity.
There were some screw-ups on my part. For example, I messed up some docking ports in the Science Wing, so Kerbals have to EVA to get past the Cyclotrons. And I didn’t have ion engine technology when I built it, so I don’t have any xenon storage whatsoever, which is important for probes.
We’re also still sitting in a ring of debris that threatens to kill us. Sure, it’s only threatening that — for the most part, it’s completely harmless. But I don’t really want to be there when some debris comes in and tears the station apart.
And finally, we have the ability to build truly colossal things now. Origin was a fun effort, a good experiment, and a great learning experience — but why settle for a station assembled in pieces from an earlier time when that was the only way?
The Next Generation
And so, Origin 2 was born.
It’s a trimmed down, more compact version of the original — at only 78m long, I was able to design and build the entire thing as a single piece. It weighs in at 244 tonnes dry weight, but once fully loaded, it’s over 1500 tonnes.
All the life support components are gone, replaced with giant life support tanks that provide enough to supply my current 11 Kerbals for over 77 years. The station can hold 42 Kerbals, though the practical limit is 20 due to the habitat rings; any more will involve taking shifts or dealing with highly unfit Kerbals.
Station orientation is provided by way of four “Vernor” thrusters at each end of the station, using fuel instead of monopropellant. This is partly because we have much bigger fuel tanks than monopropellant tanks, but also because it gives us twelve times the power per thruster compared to regular RCS.
The station’s 90k units of xenon and giant ion engine allow us to boost up from a 1000km orbit to a 1500km orbit, safely clearing ourselves from the 1000km debris field — albeit costing half our entire fuel tank for just 180 m/s delta-v! I’ll definitely be holding myself to a strict “no debris” policy from this point onwards.
The total RocketPart capacity is about 353 tonnes — enough to build the entire station over again, if need be — and ensures we won’t need to resupply too often.
The nuclear reactor provides power for the entire structure, and the station has enough nuclear fuel to last almost a thousand years(!) at the current load.
And finally, the habitat wheels are counter-rotating, meaning we’re no longer weirdly violating the conservation of angular momentum or whatever. Yay proper physics!
All in all, the new station is a great success. We can build ships there, send them on missions, return them to the station, and (in most cases) transmit their science back to Kerbin, without the numerous risks of atmospheric ascent and reentry.
At this point, there’s only two things we’ll need the Space Centre for: Hiring and lifting new crew, and sending up the occasional resupply shipment.
So what’s our next move?
With our new ion engines, I think it’s time to send probes to other planets. I do plan on landing actual Kerbals on the Mun and Minmus, but getting to other planets will take a while, so I should get that started first.
We should also probably set up a small satellite network to ensure that we can control our probes at all times. Since the space station serves as a command point, all we really need are two more satellites in the same orbit, offset by 120º each. That should ensure they’re always in view of the station.
Either way, it’s looking like we’ll be launching a lot of small craft in the near future, which will be a welcome relief from hauling all these giant components around. Onwards!
In our next episode: It’s probe time!
I've chosen not to host a comments section here, but please feel free to contact me directly if you have something to say. I'm happy to hear what you think — whether it's about an article in particular, or just the site in general.