Achiving Orbit: Continue burn or wait till ap ?

[Laie]

I don't even know if it's (practically) possible to have one continous burn until 75 km :X

It is possible, and thanks to throttling it's not as difficult as you may think. Just try it: within ten launches or so you should be able to "almost" hit the desired altitude. Getting it perfectly right requires a lot more skill, or luck, or a flight computer.

But then again, even with maneuver nodes and coasting to Apoapsis I'm usually off by 500m.

- - - Updated - - -

Try 1: Simply entered "2500km" as target orbit, hit enter and enjoyed the show

Try 2: Launched to 100km, but disengaged ascent autopilot during coasting phase and created 2 new nodes to get up to 2500km orbit (change Ap to 2500km at Ap, circularize at Ap)

That depends a lot on vessel and launch profile (which I assume was too steep). Another way to try this: go from 100km to 2500km orbit in two maneuvers (raise in one burn, circularise) or do it incrementally (e.g circularize every 800km).

[impyre]

Most answers hit near the truth here. There are two main considerations:

A) Losses due to atmospheric drag.

and Losses due to inefficiency.

Concern A is pretty self-explanatory, the longer you spend in atmosphere (and the faster you try to go in atmosphere) the greater your losses due to drag.

Concern B is less self-explanatory. The oberth effect exists because burning when you're physically moving faster is more efficient (which is why maneuvers have greater effect and require less dV when performed at points on an orbit that are lower in the gravity well). If you're launching and always accelerating... then you are *always* burning when you are moving the fastest that you can be moving. As soon as you cut the engines you begin to decelerate due to gravity, it is this deceleration that causes you to have to burn more dV to circularize. This means that going to orbit with one long continuous burn is more efficient... cut down on throttle if your apoapse gets too far ahead, but keep burning. Always control apoapse with throttle instead of pitch when possible, as burning in any direction other than prograde will create losses due to inefficiency. It's a common tactic to burn towards the ground a bit to keep apoapse low while still adding to orbital velocity, this works... but can cause excessive losses due to inefficiency and drag.

The optimal flight path for a given rocket is a practice of minimizing the sum of both of these loss types, while still being able to attain orbit. The optimal path depends largely on the rocket's TWR profile during flight... which itself depends on types of engines used and staging sequences. This is a problem which cannot be solved with a closed formula, it must be solved analytically for each rocket... which makes it a real pain. The *best* way to go (in my opinion) is to experiment with rocket builds that you like to fly (given TWR, payload fraction, staging, etc) try to find what gets you to orbit with the most dV remaining... and follow that build-type and launch profile combo for all future launches.

I personally like to use three-stage hybrid-staged rockets with the "core" being carried up to orbit and LFO booster stages which get shed around 25km. My rockets generally keep TWR around 2-3, and I keep throttle at around 2 (but it's not uncommon for me to have TWR of around 3-4 at launch)... I use max throttle until I start my gravity turn, I like to get out of the thicker air as soon as possible. I've found that for my rockets (which tend to rely on aero for attitude control below 25km) a slightly steeper launch profile works best. I go up to about 150m/s at around 4-5km before starting my turn, and then it's really quite slow. I hit around 70 degrees at 10km, and around 30 degrees at 30km. So it's a bit of a sharper turn than I used before 1.0. I've found that for most of my rocket builds, with the new aero, drag (even at 25-30km) can still pose significant losses... and all my shallower turns (FAR style) have produced less than desirable results. Once above 30km, I just point at horizon and throttle down, trying to add to orbital velocity without pushing apoapse too high.

TL;DR, you have to play around with it to find what works best for you given the way you build and fly... trying to force a specific launch profile on a rocket can be wasteful and hazardous to your kerbals.

[Old Foxboy]

"Warranty is void if you press spacebar"

[Falkenherz]

I am still not able to do reliable efficient orbitals. That may be due to the fact that every rocket I launch is a different construction, with according different drag and gravity turn rates. So, what I end up doing is:

- at launch, turn by 5-10 degree; I have to do this before my boosters pick up speed and prevent effective steering;

- at 10km, turn to 45 degree;

after that, I gradually turn more so that I approach an apoapsis of 73km. Once I have reached this, I turn horizontal and even below, just in order to keep the apoapsis stable.

Currently I experiment with finding the kind of "suicide burn" so that I shut down enginges until as close at apopapsis as possible to still be able to do a ~horizontal burn for circularisation. If I turn back on too late, I pass apoapsis and have to waste more energy by turning back upright around a 45 degree angle.

Unfortunately, my results are very mixed, sometimes not even achieving orbit despite my vessel according to KER having a dV of 3500m/s.

Edit: My difficulties may have to do with what impyre just explained above (ninja´d). Did not know about the "Oberth effect". This game truly is a learning experience!

Edited May 8, 2015 by Falkenherz

[SanderB]

With a low sea level TWR rocket (between 1.2-1.5) you want to be going 100m/s when you're 10° pitched, 200m/s when 20° and 300m/s when 30° and continuing to pitch over to horizontal as you ascend above 15km and the potential for flipping your rocket is out of the way. Using guidelines like this i always get to orbit with less than 3.4km/s dV, unless I'm trying to launch something that's very draggy or unaerodynamic.

Edited May 8, 2015 by SanderB

[nikolay-spb]

it is better if you wait till AP because in atmosphere you waste more fuel because of drag, yeah drag is very little on >= 40km but still it is more efficient if you wait till AP or at least >= 70 km

[Falkenherz]

I found this a very interesting thread so far and looked a bit more into the basics of supersonic speed:

- increased drag (relative) between Mach 0.9 - 1.2

- decreased drag (relative) from Mach 1.2 onwards, however, also decreased lift-to-weight ratio

- RL launch vehicles avoid supersonic speed before reaching 30km because of inefficiency due to drag (would be ~20km on Kerbal)

The Oberth effect means that accelerating at higher speed is a more energy efficient use of fuel, i.e. leads to more deltaV.

The question is now at which heights *and* speed the drag is worse than the beneficial effect of both supersonic and Oberth effect, i.e. the real question seems to be not so much the angle of ascend, but at which height to attain which speed. However, for the former I still remember basic geometry of a triangle, each vertige being the shortest way between its points.

So, my learner´s theory is:

- for angle of ascent, make sure apoapsis is reached via a smooth altitude velocity (i.e. no aprupt changes); however, the apoapsis when in orbit should be the end point you want to aim your heading for.

- for speed, make sure to reach supersonic not at a too low altitude, but soon enough to still profit from lesser drag and Oberth effect.

- The former would then be dependend of the latter.

I wonder if 20km is the right hallmark to aim for supersonic?

My early tests with single "Hammer" boosters were probably deceiving: Accelerating with a TWR of from the start > 2.2 caused the highest apoapsis, higher TWR did not increase it further. I do not remember when the test vessel reached which speeds, what I probably should have looked out for.

Edited May 8, 2015 by Falkenherz

[Wanderfound]

For the sake of heat, controllability and minimising drag, you usually want your launch TWR to be <2. Exactly how steep or flat a profile is ideal is going to depend upon the rocket; high-drag and low-TWR ships want a steeper trajectory than low-drag or high-TWR ships.

Poorly controllable ships also call for a steeper launch, but that's more about avoiding disaster than increasing efficiency.

[nikolay-spb]

You can make TW ratio even less than 1, then add boosters, reduce their power so they burn longer, and burn along with main engine, in the end when they burn out, you will be at 10-20km and then you dump them, but your TW is >= 1 (because engine spent some fuel) so you continue flying. I found that strategy pretty efficient.

[Orange_Slime_]

I just use a load of SRB's on the launchpad to get to orbit. It was amazing

[Montag]

I have been targeting TWR around 1.5 at launch. Faster than that gets way past terminal velocity. I do try to be at 45° by 10k m. I think trying to keep tipping over while maintaining a constant time to AP - say 45 to 60 sec until AP is the desired level may be a good way to go?

So I guess I am agreeing with Wanderfound. I don't know how much a penalty you have if you shoot for an initial 75, 85, or 100k m AP for the initial burn.

[EtherDragon]

I want to add to this post the importance of a smooth consistent (and if possible, control free) gravity turn.

Note: If you have to put in a lot of manual input to turn, it's not a gravity turn - it's maneuvering.

A true gravity turn uses gravity and atmospheric drag to turn your rocket for you, no control input needed. The reason this is valuable is because you are guaranteed to have the least air-resistance at all times. This actually requires you to go slower than you are accustomed to at first. Caveat: This works best with tail-fins of some kind.

Steps for a great gravity turn:

1. SAS, Full Throttle, and Go up! (T, Z, and Space bar)

2. Before you reach 100m/s velocity pitch your rocket about 5degrees off of vertical. This is the only manual input you'll do until the circularization burn.

3. Turn off your SAS, let your tail fins do their job. Your rocket should keep itself prograde at all times.

4. Throttle back, give your rocket enough time to get to 45degrees pitch before reaching 15km. If you're at 50degrees at 10km, that's perfect.

5. Throttle up! But not too much, keep your speed below terminal velocity. Your rocket will continue to gradually pitch over following the aerodynamic flight path.

6. Once you reach 40km altitude, you can go full throttle as the atmosphere is no longer in your way, and push your apoapsis where you want it to go.

7. Shut down when apoapsis looks good.

8. Circularize like you've done a million times before.

with this technique in action! Note, the Stock Kerbal-X (from Sandbox Mode) is capable of landing on and returning form the Mun using this technique, in the old 0.90 soup-o-sphere, it was impossible.

[Pax Kerbana]

You can make TW ratio even less than 1, then add boosters, reduce their power so they burn longer, and burn along with main engine, in the end when they burn out, you will be at 10-20km and then you dump them, but your TW is >= 1 (because engine spent some fuel) so you continue flying. I found that strategy pretty efficient.

How I get up pretty much everything in career mode. I don't know if it's the most fuel efficient but it certainly is the most cost efficient. Also lots of fun riding bundles of SRBs. Do you alter thrust in pairs and drop them 2 x 2 (or 2 x 9 )?

It is very, very difficult to do by hand without producing a completely misshapen orbit that would cost a load of extra fuel to boot. Even veteran players and autopilot addons generally don't bother.

Well I am definitely a veteran player so I agree with this statement. Lets be honest, KSP roots are free... If I don't feel like making a nice rocket with good TWR (or have difficulty with making one fit the payload) Ill try for the continuous burn but almost always a have at least 30 seconds of coast. Otherwise burn for target orbit up to 300k with about 2/3 dV and then coast. Long burns are hard to put exactly where you want them... Much easier to punch it at apoapsis.

[Falkenherz]

I did not understand that booster thing. I had 2x2 hammers, but from then on it seemed to become inefficient?

[Snarfster]

I usually go straight up until I reach either 100 m/s or 2500 m. Whichever comes first.

Then a (gentle) turn aiming for around 45% at 10000 m altitude. Basically never move the "dot" outside of the prograde circle.

Mostly horizontal at about 35k altitude, continuing the acceleration. (trying to keep my time to apoapsis constant at about 30 seconds to 1 minute).

If I manage to pull it off correctly, the circulization burn is anywhere from 75 to 200 m/s.

Quite often just holding "prograde" after the initial turn will do the trick. (I've noticed it takes a bit of practice to get it right.)

(Before I forget, I keep the TWR for the first stage at 1.3, once at 30k I try to aim for 1.6 to 2.0)

[Warzouz]

From what I've understood :

If you fall to horizon too fast

- noser to heavy/long

- TWR too low

If you stay too straight/high

- bottom to heavy/long

- Too much fins at the back

- TWR too high (reduce throttle)

Am I right ?