Ways to store more fuel?

[Andrew Ridgely]

Hey everyone!  I have a quick question regarding fuel.  It seems that every time I build a rocket, I somehow never have enough fuel.  And with each fuel tank I add, it becomes slower and heavier, and overall harder to control.  Plus, it doesn't look good.  I tried replicating an Apollo Style landing on the Mun, but even the lander ran out of fuel   My designs must not be efficient enough - I just started playing KSP a few weeks ago, so I'm still learning haha - any tips on better designs?

Thanks!

    -Andrew R

[Stargate525]

Helps if we have pics.

In general, if you're adding fuel tanks and you're not getting farther, your thrust to weight ratio is probably off. If it's real sluggish off the pad, you're losing a lot of fuel keeping your rocket moving against gravity.

[Aegolius13]

3 minutes ago, Andrew Ridgely said:

  And with each fuel tank I add, it becomes slower and heavier, and overall harder to control.  

As you say here, just adding fuel is often not the answer.  If you haven't seen it yet, you may want to look into a thing "delta v." Delta-v is essentially the "currency" of spaceflight - a given maneuver will always take the same amount.  So your objective when designing is to give your ship as much delta-v as possible.  And your job when flying is to consume as little as possible by creating smart maneuvers and executing them precisely

Delta-v is intimately connected to something called the Tsiolkovsky Rocket Equation, which is a mathematical way to calculate how much you can get done with a given amount of fuel, mass and engine efficiency.  (No math is needed in the game - a mod like Kerbal Engineer Redux can do the math and show you delta-v a given stage can produce.)  Anyhow, you may see references to something called "the tyranny of the rocket equation."  In essence, this means just what you said above - just adding fuel can be counterproductive, because fuel is heavy and it takes more fuel (and potentially bigger and heavier engines) to move that fuel to where you need it.  It's a huge engineering and design problem, and the reason we needed something as massive as the Saturn 5 just to put a dinky foil-covered lander pod on the Moon. 

So if you can't get further just by adding fuel, what can you do?  This is one of the major challenges of the game, but here are a few guidelines.

-Lighten everything you can, especially the stuff at the top of the rocket.  Payload weight exponentially affects the weight of every stage down the line.  

-Effective staging is crucial.  Dropping stages lets you lose dead weight of spent fuel tanks, plus the heavy engines that were needed to propel that fuel.  If you're building an Apollo replica, you've probably already done some staging, but it takes a lot of trial and error to figure out how big to make each stage.

-Generally speaking, when designing a craft, work backwards from the top.  Put together your payload, and figure out how much fuel and what engine you need the final stage to do its part.  For example, for a Mun lander you'd need enough delta-v to get from Munar orbit to the surface and back.  And you'd need enough trust not to crash into the surface.  Then figure out what kind of stage you'd need to get that lander from Kerbin orbit to Munar orbit.  Then figure out how to get that whole thing from the launchpad to Kerbin orbit.  Note that your stages don't necessarily need to break when you move from one action item to another, though. 

-Using the most efficient engines possible.  As I mentioned above, engine fuel efficiency (called Specific Impulse) is one of the key variables of the rocket equation - but it's not the only one.  More efficient engines make everything better, but you have to balance this against the amount of thrust you need, the engine's weight, cost, and other factors.  For example, nuclear engines are extremely efficient, but they're heavy enough to not be worth it for very small ships.  Terrier and Poodle engines are some of your best bets for general purpose, Kerbin/Mun/Minmus vacuum work, but they're not good in lower atmosphere.  I've noticed that lots of new players (including me) used way bigger engines than they needed, adding a lot of extra weight.  But that may or may not be you. 

-When all else fails, you can add MOAR BOOSTERS to the final stage, so what was originally your launch engines and fuel can start later.  But again, this stuff works exponentially, so this technique may not work for very long without getting into ridiculous size and cost.  

If you want to post any screenshots of your existing designs, folks would be happy to give you tips on small tweaks to make your ships go further.  

[Streetwind]

I have a sneaking suspicion that spacecraft construction is not the only thing holding you back. You may also need to improve your piloting ability. It is entirely possible to take a lander which has enough fuel for three consecutive Mun landings and takeoffs, and run it out of fuel in a single, poorly flown landing approach.

Of course I haven't seen you fly, so this is just speculation on my part, but if you've already exhausted your options in terms of building bigger rockets...

[ForScience6686]

I would start with your payload.  Make it as light as possible.  Do you need to dock in space?  If not, dump rcs ports and drain the command pod of any mono propellant.  How many legs are you using?  3 should be sufficient for a small lander.  Put drop tanks in the outside of the lander, with the legs.  Drop the tanks upon your launch to return to orbit.  That helps reduce the fuel needed to return.  If you ever have the opportunity to drop empty tanks then do it.  Lastly, srbs are perfect for launch.  They save your liquid fuel rockets until higher in the atmosphere where they have better thrust, plus they are cheap.

 

Of course, this is all based on my experiences with the game.  People design and fly differently, and we'd have to see what your working with.  In my early time with the game I often over built rockets.  I eventually figured out I could do the same thing with a third less rocket.  This game well eventually make you an engineer if you keep at it.

[Hupf]

7 hours ago, Streetwind said:

I have a sneaking suspicion that spacecraft construction is not the only thing holding you back. You may also need to improve your piloting ability. It is entirely possible to take a lander which has enough fuel for three consecutive Mun landings and takeoffs, and run it out of fuel in a single, poorly flown landing approach.

Regarding the "how": look up how to do a suicide burn. Essentially on non-atmospheric bodies the flight profile still matters due to gravity losses. You can very easily try this out by just hovering over the ground: the fuel will be depleted soon. The same principle applies if instead of hovering, you try to make a "slow" descent (i.e. constantly keeping your vertical speed under some maximum). Or, to put it yet another way: if you come to a full stop (also vertically) 1km over the Mun's surface, you will accelerate again and have to burn yet again before impact, wasting fuel.

What you want to do, therefore, is to burn as late as possible with maximum thrust. This is however quite hard to time and perform correctly, with the risk of either coming to a halt too soon (wasting fuel) or too soon (lithobraking)

[Palaceviking]

Insanely huge rockets....done

[CSE]

Welcome to the tyranny of the rocket equation.

Numerical readouts from the Kerbal Engineer mod may help with the design, in particular to find how much delta-V a booster stage contributes, and what your TWR will be. Design for a first-stage TWR in the 1.3-1.6 range. Different players will have different preferences, but in general when building the largest rocket your current technology allows, you should be at the lower end of this range.

A readout such as KER will also make it easier to fly your designed-in launch profile: most rockets don't have much flexibility to fly different launches, there's normally one ideal ascent for a given rocket, and not much tolerance for flying outside this envelope. KSP is replicating a real-life problem in this respect. I also find that MechJeb's non-autopilot tools allow me to fly a more precise ascent than the keyboard alone, and for some of my rockets this makes the difference between orbit and a rapid unscheduled return to the ground.

Delta-V numbers from KER will also help people give you more specific advice. There's loads of good stuff above about about reducing the payload weight, solid rocket boosters for the first stage, and so on. If you can give a breakdown of your delta-V expenditure (how much at launch? in Kerbin orbit? in Mün orbit? and on the Mün?) then people can look at a particular stage of your flight and see what they can suggest. For me, getting the ascent right and consistent took some practice, but the single biggest step I made was changing how I thought about landing on airless worlds: lower the periapsis of your orbit as low as you dare - 5km, maybe? - and come in mostly sideways rather than down. The lower you are in your orbit when you start to decelerate to sub-orbital, the less fuel you will use. Have a look at some YouTube players and, ignoring their rocket design, see how low they are when they fire the lander engines, and how they're skimming sideways over the Mün at hair-raising speed.

19 hours ago, ForScience6686 said:

How many legs are you using?  3 should be sufficient for a small lander.

Coming in low, fast, and sideways for a fuel-efficient landing, I found it helpful to stay with 4 legs until I had several successful (as in, can use the rocket again ) landings. The extra leg gives a lot more stability at touchdown, and has less of a preferred direction on slopes, and I didn't always get my horizontal velocity zeroed perfectly before the ground arrived. Once I got more practiced at it, then I went to tripod landers. I think they do take a bit more skill.

Edited September 6, 2016 by CSE
typo