Search the Community

My Munships tend to fall behind the ascent profile around 10-12 km. So at that point I\'m not worried about air friction, and start the gravity turn. Not much, just a few degrees at a time until the air gets thin. I know it sounds too early to most of us, but doing this seems to pay off. The trick is to start the turn slowly at first.

For this part, it is simple. While sitting on the surface of Mun, think about where Kerbin is located. If you are on the 'face' of Mun, Kerbin should be somewhere in the sky. In this case, you should go east-ish. Conversely, if you are on the 'back' and Kerbin is not visible, go west-ish. Both of these cases have the effect of cancelling Mun\'s orbital velocity, so that you fall back to Kerbin once clear of Mun\'s SOI. It helps to adjust your trajectory to compensate for latitude. (By the way, SOI is Sphere Of Influence. Have you already learned about this? It\'s not normal astrophysics, but more like Aristotlean astronomy. Your flight path is affected by gravitational attraction from only one body.) To answer your question about throttle optimization, it is a dilemma between fuel wasted hovering and exponentially increasing air friction. Historically, this is named the 'Goddard Problem'. Basically, build your rocket to be powerful enough for a Thrust-To-Weight ratio of 2:1. Use Closettes table of terminal velocity as your ascent profile. (Thanks UmbralRaptor! Copy this table and staple it to the inside of Jeb\'s visor!) Which brings me to the gravity turn. It\'s another dilemma like the Goddard Problem, but this time in 2 dimensions. You need horizontal velocity to reduce the effect of gravity. But there is also that pesky atmospheric drag, so you need altitude to escape drag. I just realized the other day that a rocket with TWR of 2:1 cannot keep up with terminal velocity past 10 Km or so. When following that ascent profile table, simply start your gravity turn when you can\'t keep up. In this way, air friction is minimized, and you get that gravity turn bonus as soon as possible.

Just a couple of points: - As the 'someone who did the math' http://kerbalspaceprogram.com/forum/index.php?topic=7161.msg106460#msg106460 I feel I should point out that the efficiency vs. ascent speed curve is fairly broad around the maximum so you need not follow those speeds exactly - plus or minus 20% is OK. In fact I fly my usual craft faster than the most efficient speed at the beginning of its launch (so as to burn off some fuel weight ), but I still end up about 20% slower than optimal by 12000m. Not a big deal. - As for 2D optimal trajectories from Kerbin to orbit, I\'ve been doing a lot of reading about optimal control theory, gravity turns, linear tangent steering etc. and it\'s not an easy problem to solve. (Even the lunar ascent profile with no atmosphere is difficult enough). For Kerbin, turn too early or too flat and you\'ll waste fuel scraping through a long path length of the atmosphere. Turn too late or too little and you\'ll get a nice sub-orbital flight but no orbit. I tend to wait until about 25km to begin my gravity turns, which is higher than many of you, but can get me into an ~85km orbit if I do it right. For a higher orbit, I just Hohmann transfer from there, although that takes more time than a direct insertion. If someone else (an aerospace engineer) can come up with an efficient launch-to-orbit rule-of-thumb as I did for vertical ascent, we\'d all be very interested to see it!

In a gravity turn, gravity does not provide the torque that rotates the ship. So it doesn\'t matter whether the parts of the ship feel gravity separately. Rather, in a gravity turn, gravity is what bends (turns) the trajectory downward. The ship still has to rotate itself to line its main axis up with its velocity.

So is there a table somewhere that shows altitude verses speed and gives an indication of when it is best to start your gravity turn after launch? I hear some say 30k is the best altitude and some say 40k. Some say it depends on your craft. Unless I\'m mistaken, the true factor is your speed. If you have a lot of speed then you can afford an early gravity turn but if you\'re going slow then you\'ll have to reach a higher altitude so the atmosphere doesn\'t pull you back down. Is there a \'optimum curve\' to work this stuff out, assuming a constant target altitude? Cheers.

But getting there is half the fun! Thanks for taking a look. Yes, I just focused on the descent phase because it\'s easier to analyze, and for most people does not involve staging at this time so the mass stays ~ constant. Approaching directly or from an elliptical orbit - not covered, and if you\'re after a particular landing spot you\'ll need a gravity turn and some optimal control theory to get the most fuel-efficient approach. Speaking of which, here\'s an interesting demonstration of that (in 1-dimension, but still a tough problem) on the Mathematica site: http://demonstrations.wolfram.com/MoonLandingSimulation/ At least the equation above gives one a ballpark delta-v to land as a function of altitude, for those who choose to enter orbit first with their landing stage and look for a nice spot. Something which I rarely do myself, but now I know that lower is better if I can get there with my trans-munar stage first. Oh, and playing around briefly with the KSP Orbit Mechanic Java tool (settings: Mun, Hohmann transfer) I don\'t see a big difference in delta-v from a large initial orbit (300km altitude) to say a 5km vs a 10km altitude final orbit. Both add an additional ~ 200 m/s delta-v. Of course almost no-one enters the Mun\'s Sphere of Influence with a circular orbit but these figures are a rough guide.

Glad you found this thread, Kosmo, because the credit for data-taking and inspiration for this is all due to you. I think a scale height of 4850+/-150m is an acceptable result we can all agree on? You could get better drag data above 30 km by dropping the command module from way above the atmosphere - that way you\'ll be slamming into the upper layers much faster, which will increase the drag force and its effect on acceleration. But above 30 km no-one really cares, since most people have made their gravity turn by then and the terminal speed becomes comparable to the orbital speed up there. (In fact that is why one should be orienting the nose to build up orbital velocity and not so much to push through the atmosphere by then. We all knew it, but your data proves it!). I\'d like to see (and I\'m not asking you, although it would be a huge favor) similar data for something with larger mass total maximum_drag factor, e.g. a command pod + empty LFE. I\'m just not convinced that the maximum_drag factors are simply additive in the model as claimed elsewhere. If they were, then one would expect a greater difference in the falling speeds of big vs. small objects - with terminal speed proportional to sqrt(total drag factor) - which I am just not seeing in practice. Since objects fall faster than their terminal speeds in an exponential atmosphere the only way to measure the actual drag on them is by calculating their accelerations, as you did.

The ASAS locks in a vector in privileged inertial gamespace and attempts to keep your nose on it. If you lock in vertical, Kerbin\'s going to rotate under you as you boost and make that inertial reference a retrograde gravity turn, so it doesn\'t surprise me that you\'d wind up heading west. The 290 surprises me a bit, since your theoretical turn should be slightly south of west. Maybe early \'hunting\' behavior by the ASAS feedback loop?

I\'ve been seeing a lot of videos lately where people are taking off and doing a gravity turn the opposite way I\'ve been normally doing them. Maybe I\'m under the wrong impression, but isn\'t the solid red line on the navball south? Isn\'t it easier to drop to the east rather than west?

ok...i made a video of me flying a straight-to-circular orbit and reaching a previously targeted altitude with a gravity turn... this also shows a new type of solid booster i made which burns weaker for longer (more realistic).... but the main thing is that it demonstrates how it's possible to perform the direct-to-orbit launch and become circularized at the target altitude before 12 minutes in flight i hope it's possible to learn a thing or two from it.... not sure if i explained it well enough (i was kinda busy with the flying and all) : :cheers:

don't let anyone tell you different - it IS possible to launch straight into a circular orbit if you can manage to fly steadily through a decent gravity turn... the general idea is this: - liftoff - * launch vertically, keep pointed as straight up as possible * keep your speed under control - you don't wanna go too fast or drag will waste all your fuel * watch the ADI (the ball thingy) and keep it under control as well as you possibly can - roll maneuver - * if using solid boosters, wait until separation before this, lest things get too unstable * roll 90 degrees so east is aligned to the 'roof' of your ship (could be the other way, but upside down feelsmore 'professional') * keep the ship steady and try to reach 10k altitude with around 170~200m/s vertical speed - pitch over - * make sure you have stability and able control of the vehicle at this point, if not, eject - for you probably won't make it then... * try and keep your acceleration (Gs) in the green area, more than that puts a lot of stress on the vehicle and throws you off course * do NOT turn of SAS - pull the nose 10 degrees or so towards east and hold it a bit * watch while the yellow mark (your velocity vector) turns towards east - watch your speed * as soon as you have downrange (lateral) velocity, you'll be doing a 'gravity turn' - gravity turn followthrough - * ensure the yellow ball keeps going down towards the horizon - if it doesn't you need a larger pitch-over angle * if it goes down too fast, you're also in trouble... this is often caused by pitching over too low/slow - eject * now, the most important part is to keep your nose pointed directly at your velocity vector (yellow ball) through the turn * this will ensure you're not wasting delta-vee in cancelling momentum you've built - gravity is making your flight path curved, not thrust * the goal is to reach 45 degrees of pitch at about 25k altitude.... adjust throttle to achieve this, try not to steer too far from the ball as you go along, since that wastes fuel - second pitch maneuver - * at 35k altitude there will be no more atmosphere, and you should be getting close to level with the horizon * use whatever vertical speed you have left to keep you moving upwards up as you accel towards your orbital velocity * if all was done right, you should be converging around a 45k apokee (don't worry, it's trick to get it right without proper instruments) * at this time, you can more-or-less figure out a 'probable' apokee... that's your orbit altitude - use one of the calculators to figure out your target speed * use pitch and throttle to get yourself to reach that velocity just at the same time as your vertical speed hits zero * the above is almost impossible - so you'll more than likely have to do some burns away or towards the ground to zero out your VS when you manage to reach the corresponding orbital speed for your altitude and be at zero vertical speed - you're in a circular orbit :cheers:

Yeah. I just need more practice, and a keyboard that doesn't suck. I actually tend to use 'precise' mode all the way up from the pad--most of my boosters are pretty flexible, so I need to be gentle while trying to guide them with the Yawmaster. The shuttle directly inserted into LEO after about 1984 (originally, they would cut off the mains just shy of orbit to make sure the ET would reenter without a retro-rocket, and used an OMS-1 burn about two minutes after MECO to acheive orbit), but an elliptical LEO of about 60x200 miles, typically. (Apogee varied depending on mission profile, but perigee was generally around 60 miles.) The OMS-2 burn at apogee was required to circularize even on the direct-insertion profile. (Interestingly, the OMS engines themselves are *not* throttleable; the only way to control the OMS thrust is deciding whether to make the burn on one engine, or both.) S-IVB ignition came promptly after S-II staging, about two or three seconds after separation. There's a film from a camera inside the S-II/S-IVB adapter on Apollo 4 or Apollo 6 that shows the two being separated on the S-IVB's ullage motors, then the J-2 firing. While not as famous as the S-I/S-II staging footage, I'm sure it's up on YouTube. Lemme check... hmm. All I found in a cursory search is this footage from a Saturn IB launch, though the staging sequence was identical between the two. (Whoever put together the footage did a horrible job of editing it, because they play the initial separation at several times normal speed, then the ignition at normal speed, then the rest of the footage at about twice normal speed...)Ironically, despite not having throttleable engines, the Saturn V *was* programmed for direct injection into a circular orbit, with the initial parking orbit for the lunar missions being a circular 100-mile orbit. (They could stay that low because they were only staying for about four or five orbits, and they DID stay that low because it increased the payload they could carry at TLI.) Don't most satellites headed into a GTO go into an elliptical Hohmann transfer, then use either a small second upper stage or their RCS to circularize it? After all, once you're in a GTO, the delta-vee to raise perigee is relatively low... Anyhow, what I meant to say was that it's not that hard to directly insert into an elliptical orbit; it's just getting into a circular orbit without a separate circularization burn that I've always thought was the hard part, particularly with the instrumentation available in KSP. (Someone once claimed to be able to do a gravity turn all the way from the pad to orbit in KSP... whoever they are, they're a far better man than I, if they can. I have enough trouble doing a gravity turn when I'm outside the atmosphere!)

Ah. Do you use the caps lock key? I try to turn it on as soon as I perform my final staging (and sometimes before). It's really helpful for when you're a bit heavy-handed. The shuttle used it... In fact, NASA uses it a lot for LEO. The only launches I've observed that DIDN'T use it were Orbital Sciences Minotaur launches (and Pegasus and Taurus use similar profiles). Did the Saturn rockets hold off on S-IVB ignition? None of the launch footage ever shows past S-II separation. And yes, it's partially an efficiency thing. My technique obviously diverges from the ideal gravity turn trajectory, and thus wastes a bit of propellant in steering the ship. Thus I've developed techniques to make a simple gravity turn work instead, but these are more difficult and require a lot of finesse on the throttle. The Space Shuttle, with it's low-thrust OMS providing the final orbit insertion, is well-suited for direct insertion into LEO, as the lower acceleration and throttleable thrust allow you to fine-tune the altitude and speed you will come out of the gravity turn at. I've found that for larger, harder-to-handle rockets, this gravity-turn/throttle manipulation is often an easier way to reach orbit than manually steering; if I find myself higher than my desired trajectory, I throttle back and allow gravity to pull the nose down; if I'm low, I throttle up and try to push higher before the nose drops. To do this, you need a stable rocket, and a good idea of what you want your ascent to look like, though. Higher orbits tend to require a two-burn ascent, though, as it's impractical to keep an engine running at such low-thrust levels through such a long ascent.