Temstar

http://forum.kerbalspaceprogram.com/showthread.php/48876-The-art-of-modular-base-building Why did you start two threads?

Oh depending on how radially different in design are your three types of probes (and thus their weight/CoM) there is a simple solution: Have say 8 radial docking ports and one nose docking port. Mount your eight probes on the radial ports at departure and leave the nose port empty. When you need to release a probe, undock an opposing pair, send one off on it's mission and then move the other one to the nose docking port so it's inline with the mothership's CoM At next destination release the probe on the nose port At next destination undock another opposing pair, release one and dock the other one on the nose port And so on...

Module base building guide, as promised: http://forum.kerbalspaceprogram.com/showthread.php/48876-The-art-of-modular-base-building

Let's have some pictures first to fire up the imagination: Mun Fuel Terminal: Prototype Mun Base: Prototype Kethane Refinary: So, how are these bases built? Let's break that question down into a series of smaller question. 1. How do I move modules around? I know some people prefer mobile cranes to lift modular base pieces into location. I prefer the "bottom loader" concept where each piece of a base is equipped with a 1.25m docking port underneath arranged in such a way so that when the landing legs are deployed there is just enough clearance for this rover (I refer to it as the "Base Crawler") to slip in underneath: A gap of around this size is acceptable. You don't have to be millimetre perfect with this technique. Once you drive the rover under your module you can really feel the magnetic attraction. Even on Kerbin the attraction between the docking ports is enough to pull the rover so that it tends to go backwards and forwards under the module without any user input. To centre your rover exactly underneath just apply slight forward motion when the docking ports are pulling you backwards and then a tap backwards when the docking port is pulling you forward. Eventually your rover will stop dead centre under the module. Then you just switch to the module and retract the landing legs. The drop will also correct small amount of misalignment. When done right the module will drop on the rover and the two will merge into one craft. You can then use the rover to move the pieces around. Should you miss up just hit G again to deploy the legs again and start over. 2. How do I line up the docking ports, both bottom and side? I firmly believe that to be called a "base", the base components need to be merged as a single craft. We can then pretend the base residence can move around inside in a shirt sleeve environment rather than spend two hours putting on an EVA suit just go pop next door. If you look carefully at the three pictures at the top of the post you notice that my bases are laid out so there is a framework of pipes or base tunnels, onto which individual modules are attached. So how do I make sure when I move these piece into position and lower their landing legs the docking ports will line up? First, the pipes: Here we see a single five segment straight tunnel in the VAB. Note the landing legs - the are place in pairs in the SPH in lateral symmetry with angle snap on. The legs are pointed directly up and so make a tangent with the 1.25m tube right where its walls are vertical to the ground. That means suppose we assume the landing legs prop up stuff at a height of X meters when deployed, we immediately know that centre of the docking port is X meters off ground when this pipe is deployed, and so any other structure where its horizontal docking port will end up X meters off the ground at its centre will be compatible with this pipe. We'll define this as "standard base height" What does that mean? Suppose we have this structure: We immediately know, without even having to test that this structure is compatible with the pipe above. So if we nudge the two so that the docking port on their ends touch they will be on the exact same height (let's ignore uneven ground and weight of structure for now) and snap together. If you have a round piece and you place large landing legs with angle snap so they point directly up and then you place docking ports on the ends, that docking port will always be "standard base height" regardless of actual diameter of the round piece. So using this simple fact we can build base tunnels of almost any shape and know they will fit together at "standard base height" without even having to test them: Now obviously you need a docking port under your pipe piece for the base crawler to dock with it to carry it around. Since we know that pipes are always going to stand up at one standard base height we know exactly how much the docking port need to reach down for the correct amount of clearance to mate with the base crawler. Above is what I use for my 1.25m pipes - one radial attachment point, one 1.25m remote guidance unit and the one downward facing docking port gives just about the perfect clearance. The guidance unit gives the pipe the controllability to extend and retract landing legs when it's not connected to anything, at least until it's internal battery is flat. Remember you want to have this downward docking port at centre of mass of your pipe, or else its going to be hard to drive around! Okay well that's easy, but what about the base modules? They're not horizontal 1.25m or 2.5m pieces, they're standing vertically! How do I align my docking ports so that the docking ports on the base modules are at standard base height? Unfortunately for the bottom docking port there is no easy way to know how much clearance you have. The only way is to test repeatedly by putting the module on the launch pad, deploy the legs and drive the Base Crawler (I put it on the runway and drive it over to the launch pad) underneath and see how well the docking ports line up. If there is too much room then you have to move the landing legs higher on your module. If there isn't enough room for the rover to drive underneath then you have to lower the attach point of the landing legs on your mobile. Repeat this until you get right. Fortunately there is an easy way to make sure the horizontal docking port are at standard base height without going to the trouble of driving it off the launch pad and testing it up against a pipe or another module: Here we have a new module I'm working on - a simple 6400L fuel tank module. I've already done the test to make sure the landing legs give the correct bottom clearance. So what's left is working out if that horizontal docking port is correctly aligned. I do that by grabbing another structural fuselage piece and attach it underneath, then put a single large landing strut on it as if I'm building a pipe. Then I do this: Grab that fuselage piece and stick it on the docking port being aligned. Then adjust your view so that you virtual eye level is exactly level with some part of a landing strut on your module. The grab the docking port and move it up or down so that the part of the landing leg on the fuselage corresponding with your alignment element is exactly flat, as the case here illustrated by the blue line. Place that docking port and there you have it. That docking port is now set to standard base height when the landing struts on the module deploys, no other test is required. You can get rid of that fuselage and save the part and be assured that it's compatible with other module base parts. 3. I did what you said but my piece still doesn't line up in the field! (I know this connection is not standard base height, but the exaggerated nature makes it easy to get the point, plus the cause is the same) Sometimes, even though you've done all your measurement and alignment correctly in the VAB/SPH, when you actually try to build a base in practice this happens. The reason is because of the technique. What's actually happening here is the weight of your module is messing with suspension on the Base Crawler (or tanker rover in this case). When you line up the docking ports for standard base height the assumption is that both pieces you're trying to connect are setting on perfectly rigid support. When you actually do this in practice this ends up not being true. If your module is very heavy as is the case here it will actually squash down on the rover's suspension and make the docking port lower by a bit. Add to this the little drop you get when you drop the module onto the rover and you'll usually end up a situation where the docking port on the piece you're trying to dock is lower than the available docking port on the base. This problem is worse on Kerbin due to higher gravity than it is on say the Mun. But even there it still exists if you're moving full 6400L fuel tanks or other similar heavy modules. Fortunately there is a simple way to get around this: What you do is park your module right up against the docking port, switch to the base and raise the landing struts on the structures (usually that means legs on pipes) closest to your new module. This causes that section to droop down a bit due to gravity. And during its downward travel if it goes low enough it will find the docking port on your new module and it will become attached. Then it's a simple matter of deploying all legs and undock the Base Crawler. I am currently working on a major Mun base, once done and I've made sure all the launch vehicles for my modules are Mun-capable and all the pictures are gathered I'll release all my base modules as a set. For now here's craft files for my Base Crawler, and two example lifters for base tunnels. http://www./view/w0p0eumhsh6r932/Base_Crawler.craft http://www./view/fzr6pucl79m32sj/Base_Part_Transpoter.craft http://www./view/5l2nls223xenc05/Base_Part_Transpoter_2.craft

You don't need a perfectly flat area and you don't need KAS. Surface base building can be done perfectly fine with stock parts, all it takes is clever engineering: I'll write up a guide on the engineering tricks behind this (hopefully tonight) and give you a link, stay tuned.

Awww duct tape, it's been on every American space flight since Gemini. And for a good reason too - it was used to create the CO2 scrubber adapter on Apollo 13 and then used on Apollo 17 again (along with a map) to build a improvised moon buggie fender after the original broke off.

You're not trying to use the versions with the proofing payload to lift your own payload are you? The root part of the booster when you load it as a subassembly should be a 2.5m decoupler. If it's anything else it mostly means you have the proof payload still loaded on top. The subassembly versions work with the Subassembly Manager, I tested them and they load fine. I'm still using these as of 0.21.

Back in 0.17 I did a manned landing and return to (then available, Pol wasn't around yet) all four moons of Jool in one go. Wrote up a huge mission report with loads of pictures too including an extensive explanation the logic behind the mission plan, but it all got eaten by the forum purge and all I'm left with is this picture: Basically, to visit the four moons I use one large mothership and three landers, once assembled and fully fuelled in Low Kerbin Orbit the mothership burns its engines to send the whole fleet to Jool. Upon arrive at Jool events happened in this order: 1. Upon entering Jool SOI, the mothership adjusts its orbit for Jool aerocapture pass 2. After first pass through Jool's upper atmosphere the ship is captured into an elliptical orbit with apapsis around Tylo orbit. Due to a spot of luck and Tylo's huge SOI to semi-axis ratio the mothership is due to enter Tylo SOI on the first pass. At this point the Tylo lander (the big one connected to the nose docking port) undocks from the mothership and manoeuvres into position for Tylo orbit insertion burn, while the mothership manoeuvres away so it dodges Tylo SOI for another pass through Jools atmosphere 3. Tylo lander enters Tylo orbit and performs deorbit burn to land on its surface 4. Mothership minus one lander makes a second aerobraking pass through Jool's atmosphere and end up with apapsis around Laythe orbit. My luck wasn't that good so I didn't get encounter on first orbit. At apapsis mothership fires engine to raise the periapsis out of Jool's atmosphere 5. Several orbits later mothership gets a Laythe encounter and adjusts orbit for aerocapture pass through Laythe's atmosphere. After several pass it settles down into a 80km Low Laythe Orbit 6. Laythe lander (the one on the port docking arm with the parachutes) undocks from the mothership and makes a deorbit burn to land on Laythe 7. When the transfer window comes up, the Bop lader (the one on the starboard docking arm) undocks from the mothership and perform a moon-to-moon transfer to Bop for landing 8. After having explored Laythe, the Laythe lander ejects the atmosphere landing kit and takes off back into LLO to dock with the mothership and refuels. Having refuelled it then makes a moon-to-moon transfer to Vall for landing 9. After refuelling the Laythe lander, mothership makes a trans-Jool injection back to Jool, performs a few aerobraking pass again to settle down into Low Jool Orbit 10. The three landers take off from Bop, Tylo and Vall and all perform trans-Jool injection back to Jool, aerobrake using the atmosphere and then dock back with the mothership 11. With the fleet reassembled mothership fires its engines for transfer home back to Kerbin When planning a mission of this kind there are a few obvious choices: 1. You're going to want to use aerocapture/aerobraking as much as possible. So you're going to want to first use Jool to slow down and get into some kind of elliptical Jool orbit, disgorge as many lander as you can so they go off on their own (saving fuel that would otherwise be used for moon-to-moon transfer or injection burn by reusing delta-V from interplanetary transfer). Generally this means at the minimum the Tylo lander since Tylo encounter is so common. 2. Once you're slowed down a bit you probably want to head to Laythe. Going from elliptical Jool orbit to Laythe orbit is almost free since you get to aerocapture again at Laythe. Parking the mothership at Laythe instead of Jool allows you to use the mothership tanks to refuel the Laythe lander, thus making it very easy to design this particular lander for double landing. Moon-to-moon transfer from LLO is also less expensive in terms of delta-V compared to trans-moon-injection from LJO. 3. For Bop and Pol, the landing requirement is very similar as both are low gravity worlds. You can get away with a very small lander for those worlds. On the other hand you need a lot of delta-V going out to them due to their wide and inclined orbit, doubly so if you are tempted to attempt both landing with the same small lander. This then calls for a powerful nuclear tug stage with a lot of delta-V for moving that lander around. Logically you want to design it so the these two parts are two separate crafts that can be assembled together. as required. Take my Gilly lander for inspiration: 4. Tylo lander design will be critical to the mission since it's going to be by far the biggest lander. If you like me and you consider it too "gamey" to have a command chair only lander for the whole Tylo lander you could still save weight by breaking down the lander into three pieces: orbital module that will remain in Tylo orbit for the crew to come back, contains habitation and is only responsible for bringing the crew back to the mothership after taking off from Tylo powerful descent stage with habitation for the crew to live in while they're on the surface tiny ascent stage using command chairs that will bring the crew back Rational being since the crew is only going to be in the ascent stage for a few hours at best it doesn't really need to be a pressurised spacecraft

What happens to the centrifuge when the ship is under acceleration? Does it de-spin and have the internal reorientated so that "down" is to the rear of the ship? Or do they force everyone out? Or is the ship's acceleration gentle enough that it's business as usual and the crew just experiences slightly skewed gravity inside? One thing about the ISV Venture Star from Avatar I like is they've thought of that. Venture Star uses a pair of artificial gravity rooms laid out as bolas on the end of long hinged booms which spin around the centre of the ship for artificial gravity when the ship is cruising. When the ship is in either acceleration or deceleration phase near a destination the hinged booms stop and fold up so that the rooms are now aligned with the long axis of the ship with the bottom of those rooms pointing in the retrograde direction.

No there were no suicide pill. Jim Lovell wrote about it in his book about Apollo 13: It will be easy enough to kill yourself in space painlessly by taking off your helmet.

I would like to see second gas planet with some kind of ring 2 -3 more dwarf planets in the same orbital neighbourhood as Dres, smaller than Dres and maybe one that's a bit irregular in shape at least one long period comet one vulcanoid asteroid, well inside Moho's orbit for crazy challenge one trans-Neptunian style object that's in similiar orbital situation with 2nd gas giant as Pluto and Neptune, maybe in a binary relationship with a large moon

Soon, within a week or two: I'm launching parts for a base to the Mun as fast as I can to prove that the launch vehicles for the parts are sound. Once I have the base assembled the parts will be released.

89% was only because of the terrible crashes back in its UR-500 infancy when it was still going through growing pains (unfortunately that doomed the Zond misson). Russians tend not to do tests with test stands as much, they'll rather just put together prototype rockets and set them off. And if it blows up then they would pick up the pieces and learn how to improve it. Even the R-7 went through a lot of crashes in its early days before it was perfected. As to why Proton has such an odd shape for a first stage, yes there were good engineering reasons for it. When N1 started going into trouble Vladimir Chelomey and Valentin Glushko wrote to their bosses complaining that N1 is doomed to fail because its huge 17m diameter first stage is far too large to transport whole and so is always going to require extensive on site assembly at Baikonur. And since Baikonur is exactly in the middle of nowhere with no real rocket factory you will always get shoddy work (recall the second N1 was destroyed because a bolt inside the tank got shaken loose and sucked into the turbopump). In response to the N1 Chelomey proposes a new rocket family called "Universal Rocket" which share many common components - particular the 4.15m diameter fuel tank which was the largest size that can be transported by rail whole to Baikonur. Each universal rocket component will be built and tested in Moscow before transport by rail to Baikonur for simple final assembly: The one on the far right is the UR-700, which is Chelomey's answer to the N1. It's actually an asparagus staged rocket assembled from clusters of the common 4.15m tanks. If you look at those different versions of UR-500 (aka Proton) beside it you can see what they did with the first stage. The first stage really needs to be wider than 4.15m, so what Chelomey did was he designed six "booster" like sections which holds the engines and fuel tanks which will be bolted onto the centre section which contains only the oxidiser tank. This design is called "polyblock" as opposed to the more traditional "monoblock" design with two tanks one of top of the other. Having your first stage built this way allows each major piece to be fully built and tested in factory before being sent by railway to Baiknour in seven major pieces for final assembly. It also allows a shorter first stage which makes stacking of upper stages easier. If then look back to UR-700 you can see that the upper stage of the rocket looks speciously like a Proton first stage - complete with the centre oxidiser tanks surrounded fuel+engine assemblies. In fact this is the whole idea behind the universal rocket series - reuse as much parts as possible for maximum reliability and least development effort. The fact that UR-700 uses no tanks wider than 4.15m also means that even with a moonshot rocket, all the major components can be built and tested in Moscow before railway transport to Baiknour - it just takes a lot more train wagon than a Proton.

A couple of mine: Prototype Mun Fuel Terminal: Here you see the base's crawler transporters moving around a landed Mun shuttle Mun shuttle refuelling The fuel terminal's refinery has a docking port designed for compatible tanker rovers to dock with it in order to bring in the crude kethane from well heads in the surrounding gas fields. With that success I've decided to expand operation. In order to keep number of parts reasonable the new refinery will be built a few kilometre away from my main Mun base which will have two launch complexes to handle ships on the surface. The refinery itself will no longer handle ships, instead it just converts kethane into refined products to be trucked over to the main base via product tankers: Kethane refinery prototype, note the crude tanker and product tanker Mun base prototype

Proton? Unreliable? Just because one crashed this year doesn't make it unreliable. In fact Proton at one stage was going to be a man-rated rocket lofting Zond spacecraft around the moon:

Aside from trying to build the landers so they're at least roughly the same mass, one neat trick is build the mothership with more tankage than the mission requires. Then what you do is de-fuel the heaviest lander into the mothership tanks so that its wet mass matches up with the fully fuelled wet mass of the second heaviest lander (which will be docked to the opposing radial docking arm), and like wise the other radial pair. Then once you've reached your destination you can fully refuel the landers from the mothership tanks before releasing them. This for example is my Jool fleet with 3 landers. The big Tylo lander on the nose (that's another trick, use the nose docking port of your mothership if you have a ship that's radically different in mass to the rest) leaves the mothership during Jool aerocapture and head to Tylo while the mothership and two landers go to Laythe. The smaller lander then go to Bop from Low Laythe Orbit while the medium lander land on Laythe, launch and refuel from the mothership orbiting above and then head to Vall. After that all three landers take off and re-join the mothership at Low Jool Orbit. Pol didn't exist back then so it doesn't get a lander. On both the trip to Jool and back I would trim the assembled fleet by offloading Laythe/Vall lander fuel into the mothership so that that lander is the same mass as the Bop lander. One other trick is to break the heavy landers into multiple stages stuck together via docking port (eg a descent stage and a docked lightweight ascent stage, or an orbital tug docking with a lander). When docking you can then break up that lander into component pieces to better distribute that weight around the mothership. At destination undock all the pieces to reassemble the craft before it heads off on its own. Here is my Gilly lander and its nuclear tug stage reassembling in Low Eve Orbit. Gilly's low gravity means I can have a tiny RCS only lander, but the heavy gravity of Eve and Gilly's erratic orbit means I need a powerful tug stage for it with a lot of delta-V. Since the combined lander+tug is much heavier than the Eve rover on the opposing docking arm what I did is install a docking port on the bottom of the rover and move the Gilly lander to that side, thus balancing things out.

For rovers in the 20+ ton range you might want to consider integrated landing equipment - either small rocket motors (R24-77 and its brother is a good one) and fuel tanks built directly into the rover or small rockets strapped to the side of the rover to be jettisoned once the rover is on the ground. The first solution also gives the rover sub-orbital hop capability in case you need to jump over rough terrain ala that scene from Armageddon. But if you were going for a Lunokhod style ramp I have a suggestion: instead of attaching the rover to the ramp by bottom mounted decoupler use horizontal mounted decoupler (so attach point at front, back or side of the rover) instead. I find that horizontal attachment suffers a lot less from "wheel clipping through ramp" type issues.

Manned Eve return mission should probably be the last goal you attempt as an KSP player in the current state of the game. It generally requires these skills: interplanetary transfer aerocapture (both to and from) docking (to meet up with a return craft, else your lander will be impossibility large) extreme engineering (to build a lander that can safely touch down with at least 8000m/s in the tank) Precision atmospheric landing (to land on the mountain top) You're better off building up that repertoire of skills one at a time elsewhere instead of fail over and over again on Eve. Interplanetary transfer and aerocapture could be practised with trip to Duna or Jool. Advanced rocket engineering could be attempted by going for a manned Tylo or Moho return mission. Docking by building space stations and you can try atmospheric landing right here on Kerbin. Only once you've ticked off that list should you attempt an Eve return mission. One way mission is a whole different ball game though and one way Eve landing is probably the easiest of all interplanetary landing missions.

I know what video you're talking about, it's one of the Little Joe II launches that was to be a test for the Apollo launch escape system: https://www.youtube.com/watch?v=AqeJzItldSQ It was wired up wrong which caused the fins to go hard over and stuck in clockwise roll position.

I thought Elysium's story was pretty poor, particularly that dodgy romance. But the world building in that movie was pretty good. However I think KSP ruined it for me. All the questionable physics like shoulder fired missiles size of a coke bottle reaching LEO, single stage to orbit VTVL shuttle and so on kept preventing me from getting immersed.

This. If you have some kind of facility to house/return kerbonaut from Munar orbit then an escape vehicle could be extremely simple - just build an open air lander and you'll be able to get 20 guys back in under 5 tons. In fact NASA had serious plans for a "DIY" Lunar Escape System for long duration Apollo missions to return the two moonwalkers back to the CSM in the event that the LM ascent stage engine fail to fire. Using the astronaut's own EVA suit as life support you could build a craft so simple that it could be built out of materials salvaged from the LM:

No at build time, anything the docking port is connected to is considered "docked" to the docking port and you get the right click undock option which will sever the connection. You only need two docking ports if you want to "redock". You could just action group the undock. Else build your rocket in a way that the docking port is visible for right click. For example if you use a 1.25m engine as your SM engine you will be able to click on the docking port without the shroud covering it: http://forum.kerbalspaceprogram.com/showthread.php/28133-0-20-Apollo-style-spacecraft-Apokee-MS Oh yeah one more hint - the engine shroud that will be connected to the docking port is non-physical so you can just ignore it and dock your CM to it regardless. But if you dig the neat look what you can do is go to the tracking station after LEM separation, then come back to your craft. The shroud will be gone.

I'm back playing KSP, at least for a while. But I promise I'll release the modular base kit and share the construction techniques this time around before I disappear again. In the mean time:

Since by Moore's Law number of transisters on CPU doubles every 18 month, if the only limitation is maximum operations per second per core then we could potentually have KSP double the number of parts it can simulate comfortably just by brutal force.

What's so bad about Long March 2F? Seems pretty comparable to Soyuz-2 to me. Yes it's hypergolic fuel for a manned rocket which is a bit scary, but Titan II and Proton both did it, and it's 10 for 10 success so far. Fair enough though if you're talking about China's lack of heavy lift launch vehicles. But that said it is perfectly possible to assemble modular space stations and even circumlunar stacks using Soyuz/Long March 2F sized vehicles.