Get into orbit efficiently?

[Spyritdragon]

Hi everyone,

Recently i've been having one big problem: i seem to need extensive amounts of fuel to get even reasonably small payloads into orbit. Even with ordered Asparagus staging, i seem to need about 20 units of fuel per single one i want to get into orbit to refuel my station. I don't see how i would even fly to Minmus here - just today i had to EVA three of my Kerbals into Munar orbit with their packs because their craft didn't have enough fuel left to get out of its suborbital trajectory (Bob's second similar experience), after i had to do a small surface deplacement to get to the position of my rover, then later pick them up with a different version of the same craft. And the first craft had even refuelled at my space station in orbit.

My usual procedure for getting up into orbit is burn up to about 200 m/s, then stay stable there untill i reach 10km of altitude, pitch over to 50 degrees east, burn untill my apoapsis is between 80 and 100 km, then wait till i get to apoapsis and burn till my periapsis is also in space. Is there a particular point of inefficiency here, or are these amounts really normal? And if so, could anyone share a craft file with me of a simple Minmus, Duna, or Eve lander (and returner), and their procedure for orbiting a ship?

Thanks in advance :-)

[sal_vager]

It'd help if we could see your rockets, maybe they are too heavy? You don't need orange tanks to reach Duna for instance.

You can see several craft capable of reaching other worlds here including one by me, and none of them needed nuclear engines or jumbo-64's, as for getting to orbit, I pitch over to 45 degrees east at 12km, then 67 degrees at 24km, then about 80 degrees at 32km.

I try to keep my craft from wasting fuel by pushing against the atmosphere too much, mainly by keeping the G force acting on my craft below 2, 1.5 is fine.

When apoapsis is over 70km I coast to orbit before rounding it out, but like all things, it takes practice.

God luck with your orbits Spyritdragon

[UmbralRaptor]

Unless you use jet engines, you need to burn a minimum of ~5 units of fuel for every one you bring to orbit.

It's possible to get more efficiency out of the ascent -- don't pitch over directly to 45° at 10 km, but begin gradually turning at ~5 km, ending up near horizontal at 30-50 km.

Ideal throttle position is somewhat complicated, and depends on craft design. If you have 3+ stages to orbit, it should be able to stay at (or very near) 100% without going too fast.

[SRV Ron]

There is a sweet spot in design and launch speed to reach orbit. Find it and your second stage is even capable of pushing a payload into Mun orbit. Example. My one design that weigh 451 tons on launch placed 51.5 tons into low, 50K Mun orbit. That included the second stage with fuel to spare.

One of the example Mun landers I tested contains enough fuel after insertion to bring the lander nearly to the surface before it is ejected thus saving precious lander fuel for landing and return home.

In short, an ideal design of fuel and power balanced by a good gravity turn will place a lot of payload into orbit and beyond. It takes moderate power to get into orbit. Once into orbit, a small light efficient engine will do the job of flying to Mun and other planets.

[tomf]

You could squeeze a little more performance out of your craft by following the speed profile given on the Kerbin wiki page http://wiki.kerbalspaceprogram.com/wiki/Kerbin

[K^2]

You could squeeze a little more performance out of your craft by following the speed profile given on the Kerbin wiki page http://wiki.kerbalspaceprogram.com/wiki/Kerbin

These are only correct for vertical ascent. Once you start to rotate, the optimal speed changes.

[tomf]

These are only correct for vertical ascent. Once you start to rotate, the optimal speed changes.

True, but for the small angles that are appropriate below 10k they are pretty good, and above that it tends to be as fast as your craft can accelerate anyway.

[Dun Kirk]

Once out of the atmosphere, you ideally should use something with high ISP like Nuclear engines ISP800? super efficient and painfully slow.

I'm currently getting stuff into kerbal orbit then a nuclear tug takes it where its needed.

[Spyritdragon]

Sorry for the late reply, still a very bad internet connection here, but thanks for the hints everyone :-). I'd never thought about the g force meter before. I'll be trying out a few of these tips and tactics, and ill upload a few of my craft files too.

Any way to upload them directly to the forums, or do i have to use a site like mediafire?

If not, i'll get some screenshots up :-).

[Spyritdragon]

This is the launch vehicle for the supply module of my first orbital station Skyjeb. It's already very very wobbly, so i can't imagine bringing anything taller very far into orbit. All SRB's fire, aided by my liquid engines to maintain the right acceleration. The short ones drop off to allow my first asparagus stage to fire, then the longer ones fall off, after which i continue asparagus staging as usual. I need to control this nicely after my ascent to get to my station without using some of the payload fuel, payload being the tank with the batteries, the monoprop tank and the tank with the radial engines. I already seem to have a tall vehicle here, so i have no idea how i would design something capable of controlled Kerbin escape, braking for a Duna orbit, then getting back into Kerbin orbit.

Below is the hopefully named Mission Minmus, only it never got past the Mun. It's wide rather than tall, since i heard somewhere friction is calculated in odd ways in KSP. As above, all engines fire for a nice ascent rate, the small outward tanks fall off (i needed those to have enough fuel in my lower stage, since my payload poodle engine wasn't powerful enough for orbit), then SRB's, then the four radially attached tank/engine systems, then central stage goes on. I got it to orbit, upon which it had to refuel at previously mentioned Skyjeb station before continuing to the moon. I arrived there, small unaccounted burn to get my periapsis outside of the mun, then retro burn to catch into orbit, landed on the mun. Made a few dozen kilometers of movement over the munar surface with it to get to a previously landed rover, then attempted to take off. Not enough fuel. I had to rescue my kerbals with their jetpacks and send the craft over again. It serves okay as an orbital tug, but something must be fundamentally wrong with my approach if a poodle engine with three full tanks can't make mun, a small munar surface covering, and back to Kerbin, starting fully fueled in Kerbin orbit. And i can't figure out what :-(.

As you can see, both ships carry pretty high amounts of fuel (or so i think), with what i believe aren't too heavy payloads, and both need their payload's fuel to actually make orbit. With the abovementioned launch methods usually i can put payloads into orbit without using pretty much any of their fuel, but that's also it. It doesn't go any further.

Any idea as to what could be going wrong?

[Viper_607]

When is it that you eject the lower central stage on the flight? In orbit around kerbal? if not, at what altitude and what Apoapsis/Periapsis. From what I can see, there might not be enough fuel in the central stack to take it far enough into orbit to avoid wasting the fuel of the lander.

Also are you doing any of the following:

Gravity turns during atmospheric ascent.

Doing a 90 Degree heading.

Doing excessive orbital correction burns.

"Brute forcing" your orbital transfer.

[esinohio]

If you are not adverse to using Mods you could grab one such as the Kerbal Engineer to help get your delta-v all worked out for your orbital tug and your Minmus lander. It makes it so much easier to build your rocket if you can see the impact each part has on the total delta-v for each stage and the total delta-v for the rocket. I'm sure there is a way to calculate all that information by hand if your so inclined. Armed with that info and maybe one of the great delta-v maps floating around in the forums there is not a single celestial body you can't visit!

Also, the KW rocketry mod has a great fuel tug/whatever you want to stick on top of it lifter type thing. Even if you don't want to use the mod just peeking at the rocket design might help you get your rockets sorted out and those creative juices flowing. I mean, why reinvent the wheel? Shamelessly copy the designs and give those creators a friendly shout out.

[NeilC]

Poodles are the least efficient engine in KSP. A cluster of 4 LV-909s gets you almost as much thrust for less weight and a better atmo ISP. A cluster of 5 gets you more thrust for the same weight. There's no reason to use poodles. It's a fairly minor factor affecting your design, just wanted to get that out of the way.

For your launch vehicle, efficiency in KSP at the moment means asparagus staging and you would benefit by using more of it in this design.

- Ditch your solid boosters, you don't need them.

- put those small outrigger drop tanks on top of your mainsail boosters, they don't need to be separate.

- Add one of the thinnest tanks to the bottom of each those booster stacks, right above the mainsail, this prevents them from overheating (KSP's heat model is messed up).

- Asparagus stage your 4 boosters.

- Test the result. Does it put your payload in orbit? If so, it's good enough. If not, do a 6-way instead of 4-way asparagus and it definitely will.

For your payload, if the LV-909 cluster isn't enough then consider ditching the central engine entirely and replacing your 2 outrigger tanks with LV-Ns placed radially on the smallest 1m tanks. Feed the central tank out to the engine tanks. Nuclear ISP is much higher, this should give you enough delta-V to get that capsule to Jool.

Edited July 8, 2013 by NeilC

[Spyritdragon]

Viper: im not really sure what a gravity turn is. Im not doing anything unusual though. My orbital transfer is pretty much just planning a prograde-only maneuver node at the right time to get my apoapsis slightly past the place where the moon will be, then once i get to my mun periapsis i burn retro till my orbit closes and becomes approximately circular. During ascent, i fly close to 90 degrees when im at apoapsis of 70km+ but i still need to raise my periapsis. I do this less with the tips earlier in this thread though.

esinohio: ill try, but im not too experienced with mods yet. I've only installed Kethane and KAS as of now, and i haven't even figured out how to use the parts (they just sit there), so it'll be a while before i get good at using mods. But once i do ill try those out :-).

Neil: Thanks for the tips :-). Ill try replacing the poodle engine and applying your tips, and see if it helps :-).

[esinohio]

No worries Spyritdragon, if you managed Kethane and KAS then you are well on your way. Using Engineer is a breeze. Its as simple as throwing the part on to your rocket in the VAB and you will be presented with some great info right away. With a little playing around you can even see how your designs will work within other planets/moons spheres of influence. If I remember correctly the parts are in the science section and weigh almost nothing.

Now as far as gravity turns I would suggest peeking at the many very well done YouTube videos out there when you get a chance.

Has done a great series of instructional videos as well as ones that are meant to be more FYI/fun type of videos. The one I linked there specifically mentions gravity turns. There are many many other instructional videos out there from other people as well. One could spend days looking at all the content this community has put up there on YouTube.

In a nutshell though a gravity turn is a launch technique that allows you to use Kerbins rotational forces to help sling you into your final orbit (Extremely overly simplified and I'm sure our resident rocket engineers just winced). When your rocket gets to an predetermined altitude (Not sure how this is determined so I copy Mechjeb and start at 5000m) you will begin to nudge your rocket toward the horizon in a nice gradual fashion. Where you point your rocket will determine your inclination. Aimed right at the equator will give you an equatorial orbit where aiming toward a pole will put you an a polar orbit.

Hope this helps!

[Spyritdragon]

I tend to launch East for that reason, yes. The one probe i've put into polar orbit i simply burned North and then a bit west to counteract the rotation of Kerbin.

My internet is horrible at the moment, but when i get the good one back i'll be sure to look at the videos :-). I'd seen Scott Manley before, but hadn't run into the instructional videos yet.

I'll download the mod and try it out, if i can get it to work. At the moment my modded parts (KAS and Kethane) don't have any context menu, something im still looking around for. Once i get it right ill install Engineer :-).

Still testing on my most recent version of Mission Minmus (hoping i might actually get to Minmus after a refuel this time). I'll report back when i next can :-)

[capi3101]

Lemme make sure I've got your payload down correctly here:

Command Pod Mk1-2 x1, Rockomax Brand Decoupler x1, LT-2 Landing Strut x4, FL-R1 RCS Fuel Tank x1, Rockomax X200-16 Fuel Tank x3, Rockomax "Poodle" Liquid Engine x1, plus fuel ducts and struts.

Correct?

So if that's right, that's 37.8 tonnes payload, 24 of which is fuel. 390 Isp in vacuum...gives you 3,855 m/s of delta-V. You need 2,480 give or take to perform a round trip Minmus mission once you're in orbit, so yeah, the design should be able to do the job.

Makes me think the problem may be more with piloting than design, especially if you're making it to orbit without having used any of the fuel in the payload. How are you doing your Mun descents? For that matter, how are you doing your transfer burns?

[UmbralRaptor]

Poodles are the least efficient engine in KSP. A cluster of 4 LV-909s gets you almost as much thrust for less weight and a better atmo ISP. A cluster of 5 gets you more thrust for the same weight. There's no reason to use poodles. It's a fairly minor factor affecting your design, just wanted to get that out of the way.

If you want to see something sad, try replacing a Poodle with an LV-T30. The resulting craft will most likely have better TWR and more ÃŽâ€V, without sacrificing too much control authority or landing leg reach.

[John FX]

Just have more than 20kn of thrust per ton on the pad and for each stage until orbit.

for minmus you need 1kn per ton for landing/launch with spare TWR.

Also when launching, try to not go over 200m/s until above 7km altitude. Try not to go over 100m/s nearer to the ground, you will save fuel losses from air drag.

[Spyritdragon]

@Capi That, and the ASAS module, clamp-o-tron, radial parachutes on the pod and RCS thrusters. But those combined don't really weigh all too much.

I've added in Neil's design tips, a different angling timing and keeping my G-forces under 2. I've replaced my poodle with 5 LV-909's, one below each of my main fuel tanks and two under single FL-T200 tanks i attached radially for that purpose. Still need a little bit of payload fuel, but i'll change it to a 6-pylon asparagus and that should fix it. With all this, once i manage to fire a rocket into the proper inclined orbit for Minmus, im hoping ill make it :-).

To get into a Mun orbit, i plan a prograde maneuver node to put my new apoapsis slightly past the Mun's orbit, around 45 degrees ahead of where it is at the moment. Usually my maneuver nodes will tell me when i'll get caught by the Mun's SOI. Then i adjust my maneuver to get mun periapsis around 50km. (Forgot this one time, had to burn laterally to get my periapsis to be higher than the surface). Then just timewarp till i get to that periapsis, then burn to close into a more or less circular 50km orbit.

My Mun descents are a bit invent-on-the-spot. After sometimes a few orbital corrections, i tend to arrive in a (slightly) inclined orbit even with a pure prograde burn when at Kerbin - not that that's a problem right now. I wait till im a bit past the other side of the mun from where i want to preferrably land, then i burn retro untill the point where i'll hit the surface is a bit past the landing point. I let myself coast, then when i'm content with the point i'm over, i burn retro to kill my horizontal and sometimes partially my vertical velocity. I usually keep my navball pointing close to the horizon though, so i don't have to worry about my horizontal velocity too much. I let myself fall down to the surface, occasionally burning to keep my vertical speed from going too high. I aim for 200 m/s in the start, go to 150 when i start to get close, 50 m/s when im within a few kilometres, and about 10 for the last hundred metres. When i get to the point where i can see my shadow (i usually aim for the bright side), i cancel out the remains of my horizontal velocity, then touch down as softly as i can manage.

For leaving i just turn up the throttle, point my nose towards Kerbin, burn till i have a 50km apoapsis, then go to a 50km periapsis when i get to apoapsis. Then i wait till im at the right position and burn Mun-prograde untill i have an escape point, hopefully pointing a bit towards Kerbin. When i reach it, i wait a small while to make sure im clear of the Mun's gravitational field, then burn retro to put my Kerbin periapsis around 50km or lower. Usually leads to me arriving at about 3000m/s in velocity. Then just use my parachutes, drag and any remaining fuel to slow down and safely land on Kerbin.

@John I'll take a look at my crafts, but i think i've got those amounts of thrust, at least on my lander. And generally i try to stick to those guidelines when launching too, but i haven't been looking at the 100m/s mark yet. Ill try that on my next launch.

Thanks for the help so far everyone