Launch to circular orbit, and intercept

[ibanix]

Hi there fellow rockteers,

I've been doing OK in KSP so far: I can get to orbit without trouble and I've made it to Mun in a fixed orbit. Horray me!

I have two questions:

1) Is there a way to get a roughly-circular orbit on the first pass from ground launch? My orbits always end up elliptical.

2) How do I intercept thing also in orbit? I can pick those things as 'target' and I get how to equalize inclination, but I don't know what the normal method is to actually achieve interception. This is keeping me from building awesome space labs.

3) I lied, I have a third question: What's the best way to get into orbit of another celestial body? For Mun I have been burning prograde out of Kerbin orbit until my new orbit hits the sphere of influence of Mun, then burning retrograde when I get close to the body. This doesn't always work. If there is a "normal" way to intercept and then establish orbit of another body, I'd really like to know...

Thanks!

Edited July 1, 2015 by ibanix

[Empiro]

Hi there fellow rockteers,

I've been doing OK in KSP so far: I can get to orbit without trouble and I've made it to Mun in a fixed orbit. Horray me!

I have two questions:

1) Is there a way to get a roughly-circular orbit on the first pass from ground launch? My orbits always end up elliptical.

When launching, once your AP is at where you want it to be, cut your engines. Then, once you're close to AP, perform your circularization burn. Burn slightly above the horizon to push your AP away from you, burn slightly below to pull it toward you. Alternatively, you can lock your SAS to prograde, and when about a minute away from AP, start burning at max thrust. If your time to AP starts increasing, cut thrust, and wait until the time to AP is lower (for example, when 30 seconds away, and then 15 seconds etc.). Repeat this process with a lower and lower throttle and eventually you'll get into a nearly circular orbit.

2) How do I intercept thing also in orbit? I can pick those things as 'target' and I get how to equalize inclination, but I don't know what the normal method is to actually achieve interception. This is keeping me from building awesome space labs.

There are multiple guides available, and different techniques. I have one myself here: http://steamcommunity.com/sharedfiles/filedetails/?id=289580294

3) I lied, I have a third question: What's the best way to get into orbit of another celestial body? For Mun I have been burning prograde out of Kerbin orbit until my new orbit hits the sphere of influence of Mun, then burning retrograde when I get close to the body. This doesn't always work. If there is a "normal" way to intercept and then establish orbit of another body, I'd really like to know...

Thanks!

That's the basic idea of it. For some bodies, you need to adjust your inclination too. You might need to burn earlier or later to arrive at the same time as your target. For interplanetary transfers, the phase angle is very important.

[ibanix]

Thanks for your helpful answers!

For intercepting another body, is there a way to determine when to burn, without having to use trial-and-error? I have to try different maneuvers until I find one that intercepts the body when it will be in right spot, and this gets really tedious.

[Chaos_Klaus]

For intercepting another body, is there a way to determine when to burn, without having to use trial-and-error?

Yes.

http://ksp.olex.biz

Or use Kerbal alarm clock to tell you when the time is right.

[ajburges]

The correct answer to one is a gravity turn. A well executed gravity turn will have you elegantly slip into circular orbit. I'm talking circularization burn of less than 100 m/s here.

Gravity turns are their own discussion. A discussion that has been had many times. Just search for the term.

[Signo]

If you are willing to add some mods you could try MechJeb too (apart from KAC that is actually everything you need, Chaos_Klaus reply is the best you could get around).

Mechjeb can provide the same graphs you could get from the Alexmoon calculator directly ingame.

[Kryxal]

Also note that once your tracking station and the building you get contracts from are upgraded, you can just drop a maneuver node on your apoapsis for a pretty good circularization burn. They're also useful for planning your transfer burns.

[Guest]

One way to approximate interplanetary launch windows without any mods is to use a small probe that is placed just outside Kerbin's SOI. You can then create, drag, and manipulate maneuver nodes to find the interplanetary encounters, much like you would do for Minmus or the Mun. While the time to node execution will not be perfect for leaving low Kerbin orbit, it will be close enough that there shouldn't be a huge difference in delta-v needed.

[ibanix]

There are multiple guides available, and different techniques. I have one myself here: http://steamcommunity.com/sharedfiles/filedetails/?id=289580294

Thank you, this was very helpful. It took me four hours to get a dock, but I finally managed it.

One thing I noticed was that during the final approach on dock (less than 1000m), it was very hard to consistently keep my target and prograde aligned (or anti-target and anti-prograde). I had to do small burns to get them to line up, but then they'd slip off as the approach continued. When I got down to the last 100m, I had to keep flipping back and forth between the two craft and realigning to the other craft so that they wouldn't meet slightly offset and not dock. Is there something wrong in my technique?

[Sharpy]

For intercept, first and foremost I suggest the built-in tutorial. It gives some pretty good hints.

In short:

0. start with the same orbital plane.

1. Put a maneuver node with an orbit that passes the target orbit a little.

2. look for the markers of target location and your location

3. move your node and adjust the trajectory (both prograde/retrograde and radial/antiradial nodes are useful, as to make the encounter markers as close as possible

4. burn, meet up.

5. when near the target, make sure your speed indicator shows "Target" (click it, above navball, if it doesn't)

6. burn retrograde (it's retrograde relative to the target!) to zero your relative speed

7. aim at the target and burn -a little- - some 10-20m/s is perfectly sufficient for a 5-10km encounter.

8. If you missed by more than 300m, jump back to point 6, repeat as necessary (or alternatively, use RCS "translate" controls to keep your "prograde" marker squat in the middle of the "target" marker, correcting as necessary during the approach)

9. Once at the nearest point, fun with docking begins...

You don't need to align the plane if your encounter happens at the ascending/descending node.

[ElWanderer]

Thank you, this was very helpful. It took me four hours to get a dock, but I finally managed it.

One thing I noticed was that during the final approach on dock (less than 1000m), it was very hard to consistently keep my target and prograde aligned (or anti-target and anti-prograde). I had to do small burns to get them to line up, but then they'd slip off as the approach continued. When I got down to the last 100m, I had to keep flipping back and forth between the two craft and realigning to the other craft so that they wouldn't meet slightly offset and not dock. Is there something wrong in my technique?

One possibility:

In orbit, your craft face the same the direction relative to the universe, which means they rotate around relative to Kerbin (or whatever else you're docking in orbit of). For example, if you're pointed prograde at one point, half an orbit later you'll be facing retrograde.

This can mean you end up having to make lots of small corrections as you try to dock. The usual suggestion is to point your crafts' ports at the normal and anti-normal directions (e.g. North and South if in an equatorial orbit) as there is no rotation in that plane - the docking ports will keep facing the same way.

[stibbons]

ElWanderer pretty much has it. Not only does your craft remain pointing in the same direction, but you have to keep in mind that you're not moving towards your target in a straight line. Instead, both you and the target are moving in circular paths around Kerbin that happen to converge.

If you're not lining up along the normal plane, that means that your velocity relative to your target will drift over time.,as will your relative velocity.

Another technique to try out, once you're within 1000m or so, is using RCS in translational mode to shift your prograde back across to the target.

[Sharpy]

One thing I noticed was that during the final approach on dock (less than 1000m), it was very hard to consistently keep my target and prograde aligned (or anti-target and anti-prograde). I had to do small burns to get them to line up, but then they'd slip off as the approach continued. When I got down to the last 100m, I had to keep flipping back and forth between the two craft and realigning to the other craft so that they wouldn't meet slightly offset and not dock. Is there something wrong in my technique?

Yes, what's wrong is that you didn't use RCS. Yes, you need to perform the "small burns" but not with the main engine but with directional RCS. Thing is, at 10km distance 170m misalignment is one degree angle. At 500m it's already 20 degrees, and escalating fast. Not to mention the shorter orbit of your target the more will the orbital motion spoil your alignment. (two bodies at the same speed but different altitude or not identical orbital plane, will move in an ellipse around each other as they circle the central body. The aim "right for the target" is a convenient approximation assuming your approach takes significantly less than an orbital period. If you take good several minutes for the approach, the misalignment is to be expected.

The final docking approach is a different game than the rendezvous approach. See this post.

Again, RCS is your friend (but also enemy; if you rotate it can entirely spoil your movement direction. Switch it off when rotating.)

[ibanix]

Thanks. I should have clarified that I was using RCS for the < 1000m alignment "burns". After I figured out to turn it off when in SAS mode...

Yes, what's wrong is that you didn't use RCS. Yes, you need to perform the "small burns" but not with the main engine but with directional RCS. Thing is, at 10km distance 170m misalignment is one degree angle. At 500m it's already 20 degrees, and escalating fast. Not to mention the shorter orbit of your target the more will the orbital motion spoil your alignment. (two bodies at the same speed but different altitude or not identical orbital plane, will move in an ellipse around each other as they circle the central body. The aim "right for the target" is a convenient approximation assuming your approach takes significantly less than an orbital period. If you take good several minutes for the approach, the misalignment is to be expected.

The final docking approach is a different game than the rendezvous approach. See this post.

Again, RCS is your friend (but also enemy; if you rotate it can entirely spoil your movement direction. Switch it off when rotating.)

[Snark]

Another hint: when using RCS, turn on "fine control" mode (default hotkey is caps lock).

Why?

In practice, it's almost impossible to ensure that your RCS thrusters are distributed around your center of mass with perfect symmetry. (For one thing, your CoM shifts around as you burn fuel.) You can get it approximately right with judicious design, but in practice it will always be off a bit.

The result of this is that every time you want to, say, thrust directly sideways, your ship will actually torque a bit, which the RCS thrusters will then waste fuel in correcting for. The result is that your craft wobbles and you waste monopropellant.

Fortunately, there's a very subtle and frequently unappreciated feature built into the game. If you turn on "fine control" mode, your RCS thrusters automatically shift into a mode where they aren't just all-or-nothing; they can fire at partial thrust, and the game automatically calculates the right thrust levels for all the various thrusters so that the thrust is applied perfectly symmetrically around CoM (e.g. thrusters located farther from CoM will fire at reduced thrust).

The result is that when you thrust sideways, you get a perfect sideways burn with no torque. This gets rid of wobble and doesn't spam monopropellant.

Note that it's not foolproof; if you have a really poorly-designed craft that has very badly asymmetric RCS distribution, such as having all the thrusters located on the same side of the CoM, then there's nothing that this sort of auto-correct can do to help you. But in general, RCS control works better in fine-control mode for this reason.

[Kryxal]

It's also useful for rotational control outside of SAS use ... some craft are TOO responsive to rotation controls (because you either have almost nothing or ten times as much).

[Empiro]

Thank you, this was very helpful. It took me four hours to get a dock, but I finally managed it.

One thing I noticed was that during the final approach on dock (less than 1000m), it was very hard to consistently keep my target and prograde aligned (or anti-target and anti-prograde). I had to do small burns to get them to line up, but then they'd slip off as the approach continued. When I got down to the last 100m, I had to keep flipping back and forth between the two craft and realigning to the other craft so that they wouldn't meet slightly offset and not dock. Is there something wrong in my technique?

One thing (I'll probably update the guide later) is that you should just be pointing your ship at your target, and move the velocity marker side to side using the IJLK keys (with fine control turned on). If you're doing burns, then it's probably going to be difficult to get the precision you need. Unless the two are perfectly aligned, then as you get close, it'll get further apart (imaging shooting at a target 1 mile away, but missing it just by a few feet, only when you get closer will you notice that you're not heading directly at your target)

It's also the case that since you're not exactly the same orbit, you'll slowly drift apart, but assuming that your relative speeds are small (<1 m/s) you'll drift apart very slowly, over the course of many minutes.

[ibanix]

Thanks for all the helpful advice!

Is there a way to disable SAS from using RCS? This would be helpful when I am doing docking maneuvers and need to align, but don't want to mess up my vectors.

I picked up the Docking Port Alignment Indicator mod, which was reaaaaallly helpful.

[Kryxal]

The only way not to use RCS under SAS is to turn off RCS ... maybe they'll give us the option one day, but it hasn't happened yet. Does Mechjeb do anything like that using SmartASS?