A Handful of N00bquestions

[Kardea]

So,

As my first forum post here at KSP, I thought I'd ask a couple of questions that I've been seeking an answer to since I started playing last week. If these have been answered before, please forgive my ignorance (and feel free to point me in the right direction!).

1. Thruster Use at Liftoff
   This is my "big" question, as it would completely change my current construction methodology: Let us suppose I have a rocket that consists of a single, central Rockomax tank + Skipper, surrounded by four additional Rockomax tanks + Mainsails. Is it a) better to burn then stage the Mainsails and then light the Skipper or light all engines at once and stage as the outer tanks burn out first? I'd LOVE some input here, as I'm not sure what the generally accepted KSP practice is. I know that when I watched Shuttle launches on TV as a kid, you can bet the shuttle's little engines were burning right beside the SRBs...
   
2. Rovers & Other Non-powered Payloads
   What's the best way to get a rover onto another planet? Do treat it like it's own "stage"? Or have people found ways to construct ad-hoc "cargo bays" to carry them in?
   
3. Load Balance (specifically, Piggybacking)
   Is it feasible to build a Space Shuttle-style piggyback setup, where you have the main lifter in the form of big engines/SRBs and a spaceplane or other manned craft "piggybacked" on the side? I've tried it once so far and, no matter how I tweaked, angled, and SRS'd it, I couldn't get the thing to not flip and crash inside of 400 meters up.
   

Again, sorry if these are dumb questions! Any help would be massively appreciated, though. This is a heck of a game!

[Camelotking524]

1. Fire them all at once and have the tanks on the outside feed fuel to the center tank. If you set the outer tanks to decouple during launch, the skipper will have a full tank.

2. I've build landers to land them and have paired them on both sides of my man landers.

3. I've seen people do it with stock parts. If you look at the real shuttle the outer engines on the orbiter are angeled, with the trust directed towards the center of mass. I've been experimenting with that and some custom engines but with no luck.

[Vanamonde]

1. If an engine is not burning, it is not paying for its own weight. Of course it's not possible or even always advisable for every engine to be firing, but you always want to minimize the dead weight you're carrying. So burn all the engines in your example at launch, but there's an additional trick most players use, which is to run fuel lines from the outer tanks to the inner one. The outer engines will run out of fuel sooner and you can eject their weight sooner, leaving the middle engine to keep going with a topped-off tank. This is the "asparagus" staging you will hear people taking about, because the clustered columns of tanks once reminded somebody or other of a thingy of asparagus.

2. There are many ways to carry rovers, and it mostly depends on the size of the rover. For a big one, it might make sense to build is as a lander-with-wheels, whereas small ones can be conveniently dropped off the bottoms of little descent rockets. Some players do like to build shells around them, but in the current version of the game, there's no benefit to this and it's just for looks. In fact, it makes the rover heavier, and is counter-productive, which doesn't mean it can't be fun to do anyway. As CamelotKing524 says, it's also possible to mount them in pairs on either side of a landing ship, which neatly avoids the problem of trying to balance an asymmetrical rover on a symmetrical rocket. But while I have used this approach at times as well, don't forget that it does mean doubling the payload weight, which requires the whole rocket to be proportionally larger.

3. It is possible to build non-radially-symmetrical ships, but it's extremely difficult, even for experienced shipbuilders.

Edited May 31, 2013 by Vanamonde

[Spatzimaus]

I know I'll be repeating what some others have said, but meh, it's not like I have anything better to do right now.

1> Usually, the best method is the "asparagus" setup: all five engines burn from the start, with the outer tanks having fuel lines feeding inward. Once the outer engines turn off (which'll happen when their tanks run dry), you stage, and now the inner tank (which'll still be full) takes you to orbit. Now, I said "usually"; there are a few exceptions to this:

On one hand, there's no point in having a ridiculously high thrust at the start, since you'll be limited by atmospheric drag. Go too fast, and you'll just be wasting energy offsetting the drag. So, depending on your design, it might be better to leave the center engine turned off until the outer engines shut off.

On the other hand, one of the nice things about outer tanks is that they give you a good stable place to attach struts when lifting an awkward payload (like a space station or big rovers). Separating them in-atmosphere is, therefore, a bad thing; an empty fuel tank has only 1/9th its original mass, so depending on your design it might be better to just leave them on, empty. (Or, don't put the fuel lines across, so that the outer engines continue to burn all the way up.)

And on the gripping hand, asparagus setups require a lot of struts and such just to stay stable, plus separatrons to pull them apart when the time comes, and so on. If you're careless with your design, it's easy to watch your separating side stages knock the engine off the central component, which pretty much always ruins your mission. If you can, it's often just better to use a smaller number of stages just to prevent the issues that can come up at those points.

2> For rovers, it depends on the size. Small unmanned rovers can be lifted just like a capsule, at the top of a rocket stack; the hard part is landing them safely, but a skycrane setup works beautifully there. BIG rovers, on the other hand, are more like space stations; if you lift them in one shot, you need to pay very close attention to how you balance them. I landed a 120-ton mobile fuel refinery/base on Mun the other day, and that was really rough. In the end, what I did for myself was make a giant (~3600 tons) one-size-fits-all SSTO booster, the Bucket, that can lift any awkwardly shaped payload into orbit (or beyond); it's shaped like the bucket it's named after, i.e. a "U" shape, so that payloads can be held inside it in a stable manner by attaching struts to the sides. It can lift 450-ton payloads to a circular low Kerbin orbit or push them almost entirely out of Kerbin's sphere of influence, so a dinky 120-ton rover was no problem at all.

Now, that being said, there are several mods that create either internal cargo bays or disposable payload fairings (a separating cone/cylinder with the rover/satellite inside it), in case you want a rocket that actually looks like a real rocket (i.e., the rover doesn't look like it's being duct-taped to the top of the rocket). Right now you just don't need that, but once we get the improved aerodynamics model you'll need something like this. The really huge rovers won't be practical to lift this way, so we'll have to find some other way to get them to orbit, but small ones will just migrate into those sorts of containers.

3> Generally speaking no, it's not practical to make an unbalanced design. ASAS can correct for very minor asymmetries, but its tolerance is very tight; designs that might appear to be balanced will still find themselves pulling to one side once the thrust kicks in. KSP is just not built for the level of precision you'd need to make this sort of thing work consistently; with very careful planning you can get an asymmetric design to stay stable, but the stock data tools just don't give you the information you need for this. And KSP doesn't have the sort of controls the space shuttle did, where the angle of its engines would automatically shift when the SRBs detached (since the angle you'd have to thrust at would now change, due to the changed center of mass and center of thrust).

You can always try it, and you shouldn't completely give up on the idea just because it's hard, but just be aware that those sorts of designs are just VERY hard to get right unless you do the math yourself outside of KSP. For instance, I had an early spaceplane with two sets of engines that carried a payload satellite on its back; once the payload separated, I'd turn off the upper pair of engines, since the lower ones would be in line with the new center of mass. It was very hard to tune these by eye, so in the end I just pulled out some scratch paper and calculated where I'd have to place engines to provide the right amount of torque pre-separation.

[NeoMorph]

I can thoroughly back the asparagus staging method. Everyone uses it as it is the most efficient.

Check out my sig for some helpful tips from some other helpful guys rather than me repeating what has already been said.

P.S. Welcome to KSP and these forums... the best thing since Apollo landed on the moon.

[Kardea]

Thank you, one and all, for your input! This is great info and exactly what I wanted to know. And NeoMorph, those tutorials look great. I've lived and died so far off of Pebble Garden's videos, so I'll have to check out the ones in your sig link for more goodness.

THAT BEING SAID!... I have another followup question:

Which is better for lifting payloads: A Rockomax with a Mainsail or a Rockomax, a 3-way adapter, and 3 LVT-30s/45s? I've seen both on the forums and I was hoping for a "ruling", so to speak. Is it a matter of simply adding up specific impulse? Or is there more of those nefarious maths at play?

[Specialist290]

The easy, vague, and generally useless answer is: Depends on the size of your payload.

The slightly more useful and more detailed answer does depend on math, but thankfully there's only really two equations you need to know (at least for starters): Tsiolkovski's rocket equation, and Newton's second law of motion. The first can be done easily on a pocket calculator that has an "LN" button; the second doesn't even require that.

Now, the LVT-45 (and the LVT-30) both have a higher Isp (370 vacuum) than the Mainsail (330 vacuum). If fuel efficiency and delta-v capability are your only concern, then the LVT-45 would be the clearly superior choice; the way the math works out, independent of any considerations of mass (for simplicity's sake, at the moment), The Mainsail can only get about 89% (330/370) of "bang for your buck" of the LVTs.

On the other hand, we also have to consider the forces acting on our rocket.

A Mainsail provides about 1500 kN of thrust. 3 LVT-30s only provide 645 kN. A Mainsail plus a full "Jumbo-64" fuel tank weigh 42 tonnes. 3 LVT-30s, a tricoupler, and a "Jumbo-64" weigh a hair over 40 tonnes. (This is before you factor in any payloads that may be sitting on top of this stack; again, we're keeping the math simple for now.) To find out if this setup will actually lift our rocket, we need that other equation.

Our rockets exert a force pushing up on the rocket ("thrust"). At the same time, gravity is also exerting a force pushing downwards ("weight"). The ratio that will tell us whether or not our rocket will lift (and roughly how quickly) is called the "thrust-to-weight ratio" (TWR), and is relatively simple to figure out. We already know the thrust; to find the weight, you just take the mass and plug in Kerbin's surface acceleration (9.8 m/s^2, just like Earth's), then multiply the two to get the force of weight. We end up with about 412 kN of weight for the Mainsail configuration, and 392 kN for the 3x LVTs.

Now, to find the TWR, you just divide the thrust by the weight.

For our Mainsail setup, 1500 kN / 412 kN ~= 3.6 , which means this thing is going to shoot up like... well, a rocket.

For our 3x LVTs, 645 kN / 392 kN ~= 1.6 , which will still lift off the launchpad quite quickly, but not nearly as quickly as the Mainsail. This might actually be a good thing in the lower atmosphere (where you'll encounter most of your drag, but that's a topic for another discussion).

Of course, that's just for the engines and fuel tanks.

If we double the masses involved (which also doubles the force exerted by gravity, because of the way the math works out) to simulate a payload, then our Mainsail now gets us a TWR about 1.8. Our 3x LVTs now only get a TWR of 0.8, which, as a number less than 1, means we are not going into space today.

So, in summary: It depends on the size of your payload

[Mephane]

I want to add another tip, while we are at it: when you are at about the height where the atmosphere indicator turns to the lighter color (i.e. lower density; this is a bit higher 10km on Kerbin, but you can start at 10km), start your gravity turn, i.e. slowly turn your rocket in the direction you want your orbit to go, which is almost always East. Keep your orientation slightly in front of the prograde indicator. Do not turn too fast, when in doubt turn slower. The effect is that while it slows your ascent, it gives you much-needed horizontal speed, on which you save a lot of fuel when you burn into a stable orbit outside the atmosphere. This is, btw, also done during real rocket launches.

Edited June 2, 2013 by Mephane

[kahlzun]

To summarise some of the above (which was well explained Specialist290): Higher Isp is ALWAYS ALWAYS better to have, as long as the rocket can lift itself against the forces acting on it.

also, regarding TWR:

When landing/taking off, you will always need at least 10kN for each T of mass (less if you're landing on lower gravity worlds, multiply the '10kN' by local gravity: on Gilly for example, (which has 0.05 m/s gravity) you only need 0.05 kN/T, whereas Eve (16.7ms) needs 16.7 kN/T (preferably MUCH higher)

[AndyMay]

Sorry for such a noobish question but can some please explain what translation is. I have no idea what it is, when to use it or how to use it.

[Tarrow]

Load Balance (specifically, Piggybacking)

Is it feasible to build a Space Shuttle-style piggyback setup, where you have the main lifter in the form of big engines/SRBs and a spaceplane or other manned craft "piggybacked" on the side? I've tried it once so far and, no matter how I tweaked, angled, and SRS'd it, I couldn't get the thing to not flip and crash inside of 400 meters up.

It's doable yes, but not the easiest of things. What starts as the heavy side (full of fuel) ends up getting progressively lighter, throwing off the thrust vector of the spacecraft. It can be overcome somewhat by creative use of the lil liquid fuel engines and action groups so you can add extra thrust on the upper / lower side as required.

Rockomax 24-77s were used for pitch up / down extra thrust. Three in group one (one on the bottom of each booster / external tank, to keep the nose up during initial liftoff when tanks are heavy) and two groups of two (the ones on top of the final stage) to help keep the nose down as the drop tanks empty. Mechjeb's handling the steering and throttle limiter while I'm toggling action groups keeping it in a straight line Almost got it into a proper orbit just needs to carry a little more fuel up there.

Edited June 1, 2013 by Tarrow

[Kardea]

Great info! So... yet another couple of quick questions:

1) Full thrust vs partial thrust (for reasons OTHER than overheating) at liftoff: is there a recognized advantage of one over another? Fuel consumption per unit of thrust per second doesn't change over the throttle spectrum, right? i.e: full throttle burns twice as much fuel as half-throttle and provides half the thrust, correct? If that's the case, then the only thing I could think of would be possibly throttling back to save fuel if you're going so fast that you're being limited by atmospheric resistance. Is that a correct deduction or is there something I'm missing?

2) Solid Rocket Boosters: is it just me or do they seem a bit underwhelming at the moment? I can't shake the feeling that they should have a lot more "oomph" to them (the larger-sized one, at least). They seem to burn out -real- quick and, while they make a semi-passable thrust augmentation to an existing 1st stage, it seems like you can just as easily build an even better booster out of a few fuel tanks and another engine. Am I missing something here, too? Sure, it's got a T/W ration of 4.08, but is that enough to warrant using 'em often?

[Vanamonde]

1) For the same reason you don't want extra engines, you don't want the engines you do have putting forth less than they're full effort. An engine lifting 80% of its own weight, for example, is ADDING 20% of its weight to the load being lifted. If you exceed terminal velocity, drag due to air resistence may be so strong that it's worth reducing thrust anyway, but it's a tradeoff, and generally you should be running as close to full thrust as possible. This is with regard to takeoff in atmosphere, I should clarify. In vacuum, it's less critical, but there are still good reasons for generally using full thrust. For one thing, it's most efficient to expend dV at precise times/locations such as periapsis, so the longer the burn lasts, the less efficient you are being. 2) Right now, there's not much advantage to SRBs. They burn out too quickly to contribute much to your whole journey. But if you look at the specs, their big advantage is that they are much cheaper than liquid fueled systems. The idea is that SRBs will be a financially viable choice when the full campaign mode of the game is introduced, mostly for a quick, hard shove to help you get up to an efficient ascent speed right off of the pad. (Which is how the shuttle used them.)

Edited June 1, 2013 by Vanamonde

[Specialist290]

Great info! So... yet another couple of quick questions:

1) Full thrust vs partial thrust (for reasons OTHER than overheating) at liftoff: is there a recognized advantage of one over another? Fuel consumption per unit of thrust per second doesn't change over the throttle spectrum, right? i.e: full throttle burns twice as much fuel as half-throttle and provides half the thrust, correct? If that's the case, then the only thing I could think of would be possibly throttling back to save fuel if you're going so fast that you're being limited by atmospheric resistance. Is that a correct deduction or is there something I'm missing?

Correct on all counts. That's also why lower-thrust but higher-efficiency rockets are often better for your lifting stage if you can get away with them. Of course, too little thrust -- even if it's enough to lift you off the launch pad -- can also be wasteful, since you're not outpacing the pull of gravity enough to justify burning that much fuel. It's a delicate balancing act.

2) Solid Rocket Boosters: is it just me or do they seem a bit underwhelming at the moment? I can't shake the feeling that they should have a lot more "oomph" to them (the larger-sized one, at least). They seem to burn out -real- quick and, while they make a semi-passable thrust augmentation to an existing 1st stage, it seems like you can just as easily build an even better booster out of a few fuel tanks and another engine. Am I missing something here, too? Sure, it's got a T/W ration of 4.08, but is that enough to warrant using 'em often?

The real benefit to solid rocket boosters is their cost-efficiency -- relative to a full stack of liquid fuel and engines that provide the same delta-v, they're much cheaper. This matters less right now, though, since we don't have any campaign finances to worry about. However, I'd imagine that when we have to start actually worrying about money (for those of us who will play the campaign, at least), cost efficiency is going to be just as big a balancing factor as all the technical stuff we think about now.

EDIT: Ninja'd by Vanamonde, who basically said everything I did in different words.

[rkman]

Sorry for such a noobish question but can some please explain what translation is. I have no idea what it is, when to use it or how to use it.

Translation is moving sideways (left/right, up/down), requires RCS (operate with keys I, J, K L), used primarily for docking.

The concept is similar to translation in geometry: http://en.wikipedia.org/wiki/Translation_%28geometry%29

[Anglave]

This thread is just about sticky-worthy. Great answers here.

So to summarize, your rocket design will depend on your payload. Ideally your rocket+payload maintains a reasonable Thrust to Weight ratio during the ascent and staging (reasonable being > 1.1 and < 2.5), and which does so with the most efficient engines you can use to attain that TWR.

You'll want your speed to be low in thick atmosphere and faster in thinner atmosphere, but you want to balance that against the fact that spending any extra time lifting against gravity is also an efficiency loss. So go as fast as you can without losing too much thrust to drag. (A curve between 100m/s as soon as you can get there, and 250m/s at 10,000m is about right on Kerbin.)

Some good general rules:

You want to shed "dead weight" as soon as you can; drop empty fuel containers and unnecessary engines as you climb.

You never want to lift an engine which isn't burning (except possibly as part of your payload).

You want to design your rocket so that you stay near 100% throttle during the ascent, without going too fast or too slow for your altitude.

A plug here for Kerbal Engineer Redux. The only plugin/mod I use; it displays the TWR and Dv of each stage of your rocket during construction, without modifying the way the game plays at all.

Edited June 1, 2013 by Anglave

[Tarrow]

Full thrust vs partial thrust (for reasons OTHER than overheating) at liftoff: is there a recognized advantage of one over another? Fuel consumption per unit of thrust per second doesn't change over the throttle spectrum, right? i.e: full throttle burns twice as much fuel as half-throttle and provides half the thrust, correct? If that's the case, then the only thing I could think of would be possibly throttling back to save fuel if you're going so fast that you're being limited by atmospheric resistance. Is that a correct deduction or is there something I'm missing?

You're bang on the money there. The power required to overcome drag increases with the cube of your velocity - so too much too soon and you can waste a lot of fuel just fighting the air. Becomes less of an issue as altitude rises and air density drops so you can afford to throttle up a bit later on. In RL it's a factor at surprisingly low speeds, being the single largest factor determining the top speed of cars, motorbikes etc.

edit: that lil asymmetric launcher got to a 75km orbit and managed reentry with a bit of tweaking. I might have to experiment further (for laughs)

Edited June 1, 2013 by Tarrow

[Kardea]

You guys rock! Thanks to your help (and Pebble Garden's amazing tutorial videos), I have thus far:

- Successfully put a manned lander on the Mun, then brought it back to Kerbal safely with crew intact.

- Successfully put a probe lander on Minimus

- Successfully refueled my orbiting space station using a drone-controlled refueling "spear", then used remaining RCS fuel to deorbit the spear, thus disposing of the debris

- And, most recently, successfully put a manned rover on the Mun, a mere 20km from my "first kerbal on the moon" flag (was practicing "targeted" landing)

Thanks a ton. Would not have known what to do without you guys' help.

Which leads me to yet another two n00bish questions:

1)

The only mod I'm using at the moment is the SubAssembly Loader (and I'm in love with it). HOWEVER! A lot of what I intend to do throughout my space program is via the use of docking ports. For instance, I would like to equip one of my larger vessels with a tiny, automated "one-shot" probe. I've built the probe and I've built the vessel. The problem arises when I bring on the Subassembly (the probe). The VAB seems to really not want to let me attach docking port to docking port, if you understand my meaning. I can do it on a piece-by-piece basis, sure, but the moment I take a construction with a docking port on one side and try to attach it to an already-placed port on my "main" vehicle, it doesn't want to "click" in place. Has anyone else figured out a way to do this? I -really- don't want to hand-build every probe I have on each craft...

2)

This one is easy, hopefully: is there a way to move forward as a rover without any pod torque? Right now, my little Munrover is rather top-heavy and every time I hit "W" to move forward, I nearly risk toppling over (not to mention losing all traction on the back legs) because the game interprets "W" as "Hey, Jeb, lean the pod forward!" Anyone know a better way? Is there a button that simply "throttles up the wheels"?

[Vanamonde]

Can't help you with 1 because I don't use mods, but for 2, the default rover controls are also the default attitude control keys, so a probe core will attempt to move forward AND pitch-up (or whatever the defaults are) on the same keystroke. Go to your settings and change the rover sub-section keystrokes to something else, and you'll stop those unwanted rotations.

[falofonos]

For 1. When you construct your probe as a separate ship place a probe core, stick the docking port you want to attach to the main craft with subassembly loader to the probe core and then build your actual probe (including some sort of control pod) "downwards" from that port , save to subassembly loader and then try attaching to your ship.

It's something to do with the boss connection, KSP keeps track of which connector was attached in which order, so the only connector you can use to put things onto an existing craft is the one which was disconnected from the main controlling pod.

If that didn't work make sure there are no bits overlapped with the docking ports on either side and try again.

Edited June 1, 2013 by falofonos

[benjiful9000]

[c;losed]

Edited July 4, 2021 by benjiful9000

[Anglave]

For 2) even simpler than remapping keys -- while driving switch to docking mode and make sure "linear" is selected. The selection is made in the lower left corner of the screen.

[Tripod27]

You guys rock! Thanks to your help (and Pebble Garden's amazing tutorial videos), I have thus far:

- Successfully put a manned lander on the Mun, then brought it back to Kerbal safely with crew intact.

- Successfully put a probe lander on Minimus

- Successfully refueled my orbiting space station using a drone-controlled refueling "spear", then used remaining RCS fuel to deorbit the spear, thus disposing of the debris

- And, most recently, successfully put a manned rover on the Mun, a mere 20km from my "first kerbal on the moon" flag (was practicing "targeted" landing)

Thanks a ton. Would not have known what to do without you guys' help.

Which leads me to yet another two n00bish questions:

1)

The only mod I'm using at the moment is the SubAssembly Loader (and I'm in love with it). HOWEVER! A lot of what I intend to do throughout my space program is via the use of docking ports. For instance, I would like to equip one of my larger vessels with a tiny, automated "one-shot" probe. I've built the probe and I've built the vessel. The problem arises when I bring on the Subassembly (the probe). The VAB seems to really not want to let me attach docking port to docking port, if you understand my meaning. I can do it on a piece-by-piece basis, sure, but the moment I take a construction with a docking port on one side and try to attach it to an already-placed port on my "main" vehicle, it doesn't want to "click" in place. Has anyone else figured out a way to do this? I -really- don't want to hand-build every probe I have on each craft...

2)

This one is easy, hopefully: is there a way to move forward as a rover without any pod torque? Right now, my little Munrover is rather top-heavy and every time I hit "W" to move forward, I nearly risk toppling over (not to mention losing all traction on the back legs) because the game interprets "W" as "Hey, Jeb, lean the pod forward!" Anyone know a better way? Is there a button that simply "throttles up the wheels"?

first off, thank you so much for telling me about this mod

Now, for the problem you're having, I'm assuming one of the little green connector orbs that let you attach parts glitched out of existence. To fix this, take the part that should have the connector orb but doesn't, and get rid of it, then use undo to get it back (if there are no other parts on that part you can just grab a new part from the menu) The undo button is a bit weird in this game, so if it doesn't work (usually if you just loaded the ship or came back to hangar from outside) reload the ship so you have the part you removed, put on a few random parts, remove the glitched part, undo, then remove the random parts.

For Q2, I don't think you can fix this. EDIT:lolnvm what the other guy who ninja'd me said proved me wrong. You could stick on some tiny "ant" engines on the top facing forward, set them to toggle when you hit a button, and activate them when tipping forward. If they're the only engines on the rover then don't even bother setting up the toggle button and just use throttle

Edited June 2, 2013 by Tripod27

[AndyMay]

Translation is moving sideways (left/right, up/down), requires RCS (operate with keys I, J, K L), used primarily for docking.

The concept is similar to translation in geometry: http://en.wikipedia.org/wiki/Translation_%28geometry%29

Thank you rkman I keep hearing it mentioned but I had no idea what it actually was.