How to get to orbit with minimal delta v?

[quasarrgames]

I've heard of people getting to orbit with less than 3000m/s of delta v, yet i always need 3800m/s or more. So clearly there's something i'm missing.

What's the most efficient trajectory for getting to orbit?

[OhioBob]

Don't get hung up on trying to minimize ÃŽâ€v; that's the wrong way to go about it. You should really set as your goal minimizing the cost to deliver a unit mass of payload to orbit. Below is a recent thread in which we discussed this; it should answer many of your questions.

http://forum.kerbalspaceprogram.com/threads/134299-Launch-Vehicle-Optimization-Test-Results

The most cost efficient launch vehicles are not the most efficient in terms of ÃŽâ€v. But who cares? Minimizing ÃŽâ€v doesn't get you anything.

Edited September 29, 2015 by OhioBob

[GoSlash27]

^ Seconded. "Minimum DV to orbit" doesn't get you anything. What you want is minimum mass and minimum cost.

Vehicles that are designed to achieve orbit with minimum DV accelerate rapidly while following a gravity turn. Because they require bigger engines to do that, they are neither mass efficient nor cost efficient.

I design my stuff to use cheap boosters (you're gonna throw them away) and light upper stages. It carries a penalty in total DV to orbit but pays off in higher payload fractions and lower $ per tonne.

Best,

-Slashy

[xebx]

Start gravity turn at 400m, turn shape 40% (MJ) or 45° at 19-20km, at liftoff twr 1.5-1.6 (non atmo), 0.7 for the last 40% is ok, 3400 m/s deltaV but it'll depend on drag/lift. (mk2 parts have lift, and mk3 parts have less angular drag than rocket parts)

Edited September 29, 2015 by xebx

[Streetwind]

For the sake of answering the question:

Minimum dV to orbit is achieved by using a very pointy, smooth rocket (for example with a specially shaped fairing) with absurd amounts of TWR, going hard sideways right after launch, and pretty much pretending the atmosphere isn't there. You'll get reentry effects all the way to orbit, and your rocket is probably going to be a few degrees from exploding due to overheating on every single part, but you'll have spent less than 3000 m/s that way. Not sure what the record is, might be as low as ~2600 m/s.

The basic gist is that engines firing in the vertical is wasteful. Every second you spend vertical, you give away 10 m/s dV without getting anything in return, just like that. It is spent just to maintain the status quo versus gravity. On a normal ascent, those gravity losses are easily 4-5 times as high as drag losses. So in order to save dV, you want to minimize gravity losses first and foremost. The way to do that is obvious: don't burn vertically. You want to turn over as soon as you can, as hard as you can. This of course means that your rocket is going to fall back to the ground... unless you accelerate so fast that your rocket doesn't have time to fall. That's why you need the absurd TWR. Now, accelerating hard low in the atmosphere does give you a lot of extra drag losses. But since gravity losses are so much greater than drag losses to begin with, the trade is generally going to go in your favor. And by making the rocket super pointy and smooth, you can mitigate the drag problem somewhat.

Edited September 29, 2015 by Streetwind

[p1t1o]

I have a fairly simple method which works every time.

Design Rocket to have a TWR of at around 2 at liftoff. Does not need to be more than 2.

Blue centre-of-lift marker does not need to be behind the yellow centre-of-mass marker, but not too far ahead. Generally, one set of decent fins will give enough stability even to quite unwieldy craft, my ascent profile is gentle enough that aerodynamic stability is not too critical.

Manage throttle to maintain a 2G acceleration at all times. Ensure stages are capable of maintaining this on staging. Stages after the first can make do with initial accelerations down to about 1.6Gs, but this will be slightly more wasteful.

Rough approximation:

(Mass of ship in tons)*20= thrust required in kN for 2Gs

Starting at about 500-1000m, pitch towards the horizon (towards your chosen heading, usually 90degreee East of course) until the heading marker in the NavBall touches the inside of the velocity-vector ring (the bright green symbol). This represents an (negative) angle-of-attack of around 3-5 degrees. Maintain this at all times whilst under thrust. Maintain 2Gs.

Cut thrust when Apoapsis is at the desired altitude, and circularise when you get there.

I design all my rockets around this profile, I'm not sure how efficient it is relative to dV, mass or cost, but it works and does seem to give me a good mass-to-orbit ability without ridiculously large lifters. Its also easy to fly and seems to work for a wide variety of designs.

I generally over-design my lifters anyway, giving at the very least 4000m/s dV to reach initial parking orbit (which is usually in the 3-400km region, not many missions require orbits lower than this), though I usually give it 4300+, this gives me plenty of margin for changing orbit characteristic if I so desire, it also gives me extra margin even if travelling to other planets. I could shave off several hundred m/s dV to save a little bit of cash, but the risk/reward doesn't seem favorable to me.

Note: if you are aiming for a very low orbit (<200km) you can push the angle-of-attack a bit more (nose down towards the horizon slightly more, stability dependent) but maintain the same thrust profile.

Note II: I use FAR.

[KingPhantom]

i build my rockets with 3200m/s. thats perfect for me.

[OhioBob]

Manage throttle to maintain a 2G acceleration at all times.

I advocate that everyone play the game they want to play and have fun doing it. That being said, from a payload and cost perspective, I don't think you should ever throttle back a liquid fueled engine. If it is necessary to throttle back, that just means the engine is bigger than it needs to be. The biggest payload that a particular engine can lift into orbit comes when you load up on fuel to the point that you bring your liftoff TWR down to nearly 1.2. You then run at full throttle the whole way. The low TWR results in significant gravity losses, but with all that extra fuel (which is cheap) you can burn longer and lift a big payload for a low unit cost.

[p1t1o]

...snip...

Yeah, I'm sure its not the most efficient route under any context, but it works very consistently with any craft I fly.

Do you not get significant drag losses at full burn all the way? Steady 2G keeps you quite near to terminal velocity for the most part (it also minimises huge thrust changes on staging, which can be a problem in more unstable craft) and a nice gravity turn - full burn all the way gives you a more vertical ascent which can make circularisation a pain if your upper stage has limited thrust.

Once you are out of the majority of the atmosphere I dont see a problem with full burn, but its just my way (with the 2G ascent and 3-5 deg. AoA, I am usually at a nice 45 deg. pitch by this point)

**edit**

Full-disclosure: I use a part recovery mod and realchutes and tend to recoup most of the hardware cost of my lower stages.

[Streetwind]

Do you not get significant drag losses at full burn all the way? Steady 2G keeps you quite near to terminal velocity for the most part (it also minimises huge thrust changes on staging, which can be a problem in more unstable craft) and a nice gravity turn - full burn all the way gives you a more vertical ascent which can make circularisation a pain if your upper stage has limited thrust.

See what I wrote above... drag losses are barely worth mentioning for most rockets. Just adding up orbital velocity + gravity losses probably accounts for nearly 95% of the dV cost to orbit. When you trade lower gravity losses for higher drag losses, you pretty much always win out (unless you are literally launching a barn ).

Also, forget what you remember about terminal velocity from pre-1.0 KSP. While it used to be the case that you could match it neatly with about 2 G acceleration back in the souposphere, nowadays it runs away from you significantly faster than most rockets can follow it. It pretty much ceased to be something you ever need to take into account anymore.

[OhioBob]

Yeah, I'm sure its not the most efficient route under any context, but it works very consistently with any craft I fly.

Doing something that you are comfortable with is certainly important. For instance, I know that the method I use to land on airless bodies is not the most efficient (neither is it the least efficient), but I'm comfortable with it. When I feel relaxed and confident in what I'm doing, the chances of success are much greater.

Do you not get significant drag losses at full burn all the way?.

If I started out with a TWR of 2, like you, then that might be true. However, by starting out with a low TWR of, say 1.2-1.4, then I'm actually flying slower through the thickest part of the atmosphere and drag losses are low. My TWR is rarely above 3 by the end of stage burnout, and by that point the air is significantly less dense.

Steady 2G keeps you quite near to terminal velocity for the most part (it also minimises huge thrust changes on staging, which can be a problem in more unstable craft).

Keeping below terminal velocity was important prior to v1.0, but with the new aero model, my launch vehicles never come close to terminal velocity, even at full throttle. I agree that stability at staging can sometimes be a problem, but I find that if I set SAS to prograde hold through the staging cycle, this keeps the rocket plenty stable. And if not, then I have a poorly designed launch vehicle and its back to the VAB for modifications.

and a nice gravity turn - full burn all the way gives you a more vertical ascent which can make circularisation a pain if your upper stage has limited thrust.

I generally have little problem getting a nice gravity turn and a modest circularization burn.

Once you are out of the majority of the atmosphere I dont see a problem with full burn, but its just my way (with the 2G ascent and 3-5 deg. AoA, I am usually at a nice 45 deg. pitch by this point)

I'm not trying to convince you to do anything different or to move outside your comfort zone. I'm just pointing out a competing design philosophy. The more ideas we can toss around the better.

Full-disclosure: I use a part recovery mod and realchutes and tend to recoup most of the hardware cost of my lower stages.

That does make a difference. If you are able to recover the cost of your engines then there is less need to design for economy. My launch vehicles are typically disposable, so I need to pinch pennies.