KSP is certainly impressive. I must get into one day.
You absolutely must! It's the game of the century IMO!
hobbs,
I'm not sure what I'm doing wrong, but that first ship I sent there had just about the fuel capacity of an orange tank.. but by the time I got in orbit around Vall, most of the fuel was gone. That's why I started building this space station, so that I could get better at docking, but also so I could have fuel in orbit that could resupply the next rescue mission - giving me more fuel to work with.
I JUST attached a 3rd part to my space station by the way - more fuel - the top part with all the solar panels. My third successful dock ever! And it was done sideways! I am pretty damn happy about it.
Any space station design tips? Anything you guys see you would have done differently personally?
It's got about 2 orange tanks worth of fuel and room for 15 (not 17 like I said earlier). Only 6 on board for now.
Not sure what I'm doing next. It should be enough fuel for the rescue mission I guess.. Now I gotta design that interplanetary stage
OK here's some tips, I hope they help.
1) Kill the overkill.
Unless you're packing multiple ion thrusters, you don't need so many batteries and solar panels. All they do is add to the complexity of your design (more things to break), slow down your computer (more parts = more physics calculations for you CPU), and add more useless mass that you then have to bring more fuel to tug along, which means less useful payload.
I know, it's always fun to go for symmetry and awesome-looking, and in Low Kerbal Orbit, you can get away with a lot of waste. But when you are trying to get out to the Jool system, every single gram counts and usually counts
against you if you don't plan accordingly.
2) Rethink the systems you use.
You see those little round RCS tanks? Ditch them. They are really only useful for craft an order of magnitude or two smaller than the one's you are using them on. While a spherical tank will always provide the best [propellant mass/tank mass] ratio for tanks of differing geometries of the
same size, the problem is that there is no spherical tank of similar size to the big grey and yellow cylindrical ones. So you're adding all these little tanks, which add complexity and mass, for very, very little return in RCS fuel. Ditch them all and replace them with just one big cylindrical tank. Even then, only add a big cylindrical tank if you really need it. I'll come back to that in a bit. One last thing on RCS, make sure you turn it off when you don't need it. You also shouldn't need it while making your burn to get to Jool. NERVA's have thrust vectoring and while you are burning them, you shouldn't need your RCS thrusters to maneuver your ship.
Anywhoo, do this kind of analysis on every part you add. Before adding a part, ask yourself what it is meant to do vs what it is capable of doing. To beat a dead horse, if you need an RCS system for a large ship to do complex docking manuevers, then using an RCS tank system meant for itty bitty ships makes no sense. Similarly, if you just need to power docking lights for a few minutes and ensure capsule power never goes out, do you need 8+ large solar wings and 10trillion batteries? You can look up in the parts list the power requirements of docking lights and SAS units per second and figure out how much solar panel you need (their generation capabilities are also listed) and how many batteries you need to run it for a reasonable amount of time. Add in a factor of safety, but don't overdo it and so on for every subsystem you have.
3) Figure out the scope of your mission first. KSP is wonderful in that it's completely open-ended. But this becomes a challenge because when you have a rough idea for a mission, without the game telling you what you are actually required to do, you end up tacking on all sorts of extraneous things that interfere with primary mission objectives.
So, for example, you want to be able to get out to Vall, land and recover astronauts and have enough fuel to get home. Do you need a 15 Kerbal crew to do that? And if you feel like you need to set up a permanent research station with 15 kerbals and rescue the stranded ones, then you need to split that up into 2 different missions as the those two goals are largely contradictory. For a rescue, you don't want to send more kerbals than you need to or else you risk stranding yet more kerbals. If you want a research station, then you don't need a landing return/vehicle for initial operations (you can send one separately later) and you also don't need a ton of fuel for a return trip (you can always send more later). By tacking on a lander and lots of fuel to your research station, you're adding complexity and it becomes unworkable. Take it in smaller steps before you jump in all at once. Plus, the game itself just can't handle huge ships, it becomes unplayable.
Finally, the problem with your current station setup that you've pictured is that you aren't going to be able to move it. It's not symmetrical, so you have no idea where the CoM is and when you thrust, it's going to wobble like crazy and make burns nigh impossible. It's also liable to break without using quantum struts or hacking it. It's also going to go insanely slow and burn a lot more fuel than you think it will. You're going to get to Vall with empty tanks, which means more stranded kerbals. If you sent smaller modules and docked them, you will largely avoid these problems. And if fuel is a problem, then send tankers out first, ready to refuel your ships at Vall.
4) Interplanetary maneuvers
You know how to do these, obviously, as you gotten to Vall already. I'll give you just a couple of tips that will make your life much easier.
So when you do your initial burn to escape Kerbin, you'll see your flight path will terminate in a little circle. This denotes when you escape Kerbin's sphere of influence and it's a very important thing to keep track of. The problem is that if you have the game on time warp when you pass this point, your trajectory will get massively screwed up. This also happens when you enter the SoI of another planet like Jool. So don't timewarp through them! When you burn from Kerbin and you set up a nice encounter with Jool, warp until just before you leave Kerbin's SoI. Then stop the warp, and just let it pass through the SoI in normal time. And voila! Your trajectory won't change, which means you won't need a correction burn to correct the change. Do the same thing when entering Jool's SoI. You will need correction burns just to get your trajectory closer to where you want it, but at least you won't have to make a burn to correct a problem that only exists because the software can't handle time warps.
Correction burns should be made as early as possible in your flight path. After leaving Kerbin's SoI, immediately make a burn to bring your final encounter flight path to exactly where you want it. This will take a long time (expect to spend 10-20 minutes playing with the maneuver nodes before you get an optimal solution) and always keep an eye on how much delta-v your burn will require. Sometimes you'll find you've got an optimal trajectory, but it requires a massive amount of delta-v. This is usually because you screwed up and used the maneuver nodes improperly. If this is the case, delete the maneuver and try again until you get a more reasonable delta-v requirement.
The important thing is that if you do this correction burn early, you get much more bang for your buck. A small amount of delta-v applied early will radically change your trajectory. If you wait until you are close to Jool, you will have to burn enormous amounts of fuel for less trajectory change. Just remember that after you've made this burn(s), to slowly creep into Jool's SoI in standard time or you'll find the software will undo all of your fuel savings if you warp through the SoI.
Lastly, avoid inclination changes if possible. This is hard to avoid if you are going to have to dock multiple craft. However, do your best to put all of your craft in reasonably similar orbital inclinations. This is because changing orbital inclinations require a ton of fuel, so if you set up your encounters to come in at some extreme angle, you might want to rethink it and go for an encounter that comes in close to the inclination that you want all your ships to be in. This may take more fuel upfront, but will save you massive amounts of fuel in that you won't have to change inclinations of all your ships while in orbit.
Okay so that's a lot to take in. Sorry this post has been so long, I just enjoy the game and helping people out. I hope it helps you! I am going to make it a big longer, however. I am going to walk you through a design process for an interplanetary tanker. I've thrown a lot of information at you, but I'd like to show you how to best apply that information to achieve your goals with minimal frustration.
Hobbsian Design Process (HDP):
Before you start building up a spaceship, list all of the things it must be able to do.
Then match systems with design requirements.
Finally, build and test the system.
Interplanetary Refueling Tanker Design
Tanker Requirements:
1)Propel itself to the destination.
2)It needs to carry a decent amount of fuel as a payload to the destination.
3)It needs to be able to dock.
If it is the 'passive' docking partner (it doesn't move much to dock, the other ship moves to it), then it only needs rudimentary RCS systems. This is why I said you may not need a big RCS tank earlier. But you probably will as just turning a huge ship like a tanker requires a lot of RCS fuel.
If it is an 'active' partner (it moves to the other ship which is essentially stationary) then it needs to have a robust RCS system with thrusters aligned with the Center Of Mass. I will design an 'active partner'. *It is possible to design
all of your ships as active partners, even if you only end up actually maneuvering one of them and the other is stationary. This provides redundancy and factor of safety at the cost of payload and added complexity. You can wind up with extra RCS fuel if one ship hardly moves; however, this benefit can also be cancelled out if you have to make extra maneuvers to move the ships (which are larger due to carrying more RCS fuel - thus requiring more RCS fuel for maneuvers) around more.
4)It needs a control module (preferably an unmanned drone control pod).
5)It needs docking lights, an antenna for awesomeness and
6) power systems to power it all.
So now we've identified it's major functions, it's time to figure out how to deliver these functions.
To propel it, the obvious choice is NERVA. Ion is too power hungry and provides too little thrust but have great efficiency. Chemical rockets have too low efficiency, though they have more thrust. NERVA is in the sweet between the two and is the standard for interplanetary travel in KSP. I would recommend no less than 2, but no more than 4 engines. <2 provide too little thrust and require long burn times. >4 require more complexity in your plumbing and make your CPU cry and don't make your burns short enough to be worth it. So let's go with 4 NERVA engines, placed symmetrically on pylons around the base of your rocket. Each engine should have it's own fuel supply - and the total of these fuel tanks should be enough for a one-way trip to Vall without dipping into your fuel cargo that's meant for refueling other ships. If you've followed my tips on interplanetary maneuvers, then probably 2 long skinny fuel tanks per engine are going to be enough to get you to Vall. We'll revisit this issue during the testing phase. Also, since it's a simple unmanned tanker, don't worry about having enough fuel to return to Kerbin. This is an expendable vehicle with one purpose and should not be designed to return.
For your payload, obviously, you are bringing fuel and you need to bring a lot of it. One orange fuel tank is a pretty good load. You could consider bringing two stacked orange tanks but that requires and even more massive launch vehicle to get into orbit and more fuel for the NERVA's to burn. One should be fine for your purposes. I would cross feed this tank to the top NERVA fuel tanks (don't crossfeed to the bottom NERVA fuel tanks or it'll drain your orange fuel tank first). I do this for safety, as if your NERVA tanks run dry, it's best to burn some of your payload just to get to Vall and have a bit left over than to have no fuel at all because you can't get to your destination. However, if you can, I would disable these crossfeed lines and only activate them in the even that your NERVA tanks run out. I'm not sure if you can disable crossfeed lines though, so that's why it's important you link to the top tanks of your NERVA stacks and not the bottom stacks or to the NERVA's directly.
As an active docking partner with a lot of mass, you need a robust docking system. For starters, it should have a 6-way docking port so that you can dock in multiple attitudes. Think about putting 2 of these ports on, one at the front and one at the back. Since this is just a fuel truck, the orientation of how it will dock isn't very important, as you will eventually undock and jettison it; this isn't a permanent station module that needs to look pretty and be symmetrical with respect to the station/lander/transfer ship. By provided 6-way ports (or 2 6-way ports) you ensure a lot of flexibility in that you won't have to perfectly align your approach for a specific port. You can just dock with whichever port is convenient. This is a big tanker, so you're going to need at least one, if not multiple, big RCS tanks. Again, ditch the smaller RCS tanks as they provide limited utility for a system this size but add to complexity. You are going to need 8 RCS thrusters at a minimum, placed around the center of mass. This will enable you to make maneuvers without entering a spin. It will be very hard to flip your ship, so you may experiment with maybe 4 thrusters place at each end of the ship to provide a moment arm for your RCS thrust to turn your ship. If you do this, place them at equidistant points from the CoM, if you just stick them at the ends, you'll find your RCS system will be unbalanced. One last thing you can do for control is by placing some small engines at the top of your NERVA stacks, facing the front of your craft - these engines will be reverse engines to slow you down as they fire forward. With the button grouping function, group your NERVA engines to be activated by a number key and the other engines deactivated by the same key. Then with another key, set the NERVAS to be deactivated by that key and the smaller engines to be activated. This will allow you to toggle the forward and reverse engines on and off. With the smaller engines, you can now slow down your craft without burning RCS fuel and without having to turn around. I would not use NERVA's for reverse thrust, you aren't going to have to kill alot of velocity or anything, these are just for fine tuning, so use small engines (a step down from the poodle, but don't use the teeny tiny square engines; they're too small).
Stick an unmanned satellite core module somewhere on the ship. Doesn't matter where, it's too small to throw off your CoM no matter where you put it. Just don't put it in the way of your docking ports. Stick it on the side or something. You're probably going to want to put on some SAS modules on your ship to provide turning capability without having to use RCS. A couple of the large modules (do they have large SAS modules?) at either end should be fine. Also, stick on an ASAS module someplace to be able to use the thrust vectoring of your NERVAS and to control the control surfaces of your launch vehicle.
Place docking lights around your docking ports, but have them grouped to separate buttons so you can turn on only the set you need. No point in lighting up the forward and aft lights if you only need the forward ones. Stick an antenna on the ship somewhere out of the way, same thing with any scientific instruments.
Figure out how much power you need to run one set of docking lights, your SAS units and your ASAS unit. Only put on enough batteries to supply power for a set amount of time - you don't want to be able to power it for 50 hours in darkness as that's unrealistic. Only put on enough solar panels to run your systems in daylight without draining the batteries, don't add more than you need. If you can run the systems in the daylight, then add a bit more solar panels so that you can run your systems and recharge your batteries as well. You don't need to worry about recharging batteries instantly, that's overkill. Also keep in mind that at Jool, the solar radiation flux is weaker, so you'll need a few extra panels to overcome that. I usually don't bother with XL panels, they're overkill and heavy. I typically just attach the simple, unmovable square panels at intervals around the tank; this will ensure you get some power generation at all spacecraft attitudes. If you are really scare of running out of juice, attach a single RTG (nuclear generator) opposite your control pod. This will balance out the control pods mass for your CoM calculation (though it's not really an issue) and will ensure you never run out of juice, though it won't be able to run all your systems on. It's more of an emergency back up.
Ok, so that covers tanker design. Before you send it off, you should test it first. Design a launch vehicle capable of putting it in orbit. I typically go with a launcher that has 8 double stacked orange tanks around a central single orange tank core, each stack with a mainsail at the bottom. I used asparagus staging, though you can do without it or use simpler staging. Place your payload on top of the single core tank. This will handily get you to orbit. If you use asparagus staging, and/or use double stacked orange tanks for the central core, this launcher will actually be able to do alot of your initial burn to Jool, saving your NERVA fuel. You can experiment with launchers though; that's on you. My philosophy on launchers is that you do want overkill. You need to be able to get to orbit without burning your tanker's fuel, obviously. But since you already have to build a huge launcher to haul it to orbit, make it big enough to be able to do some of the Jool-injection burn. Using such a launcher will save your NERVA fuel for when you really need it. A launch vehicle probably won't be able to do the entire Jool-injection burn, but anything that it can do is better than nothing if you used a launcher that is only capable of putting your tanker in Kerbin orbit.
Anywho, get it into Kerbin orbit to do a full system check out. Go to the night side and see how fast your systems drain the batteries. Go the light side with nearly drained batteries and see how fast they charge. Check out your RCS system, make sure it's adequate to change your space ships attitude. Send up a docking ship (preferably a second tanker) and have them dock to get a feel for how your ship handles during docking and to iron out any kinks. You really, really want to do this before you go off to Jool as you'll discover all of the flaws of your ship before you go through all the work of going to Jool only to discover a mission-ending flaw.
Finally, send a tanker on a Jool trajectory. Expect to lose this ship. If everything works well, then you can use it to refuel your other ships at Jool. However, I promise you this first attempt will end in disaster. You'll mess up a burn or you won't have enough solar panels or you'll find you don't have enough fuel to make it (which means you'll have to add more NERVA fuel tanks to avoid dipping into your payload tank), whatever. This test run out to Jool will verify your design and allow you correct flaws *before* you need the system to function perfectly. It will also provide you with practice in using the system.
After all that, send out a fleet of tankers to Vall or Jool to await the arrival of your rescue mission and your scientific station.
OH GOD I JUST WROTE A MASTER'S THESIS
w00t w00t
