Burn Planning with LV-N for High-dV Maneuvers

[Srpadget]

My Kerbals are finally getting ready to go interplanetary in a big way. That obviously means nuclear propulsion. But 'interplanetary' means I'll be doing larger-dV single burns (>2000 m/s in some cases) ... and with fuel-efficient but low-accel craft? That means long burn times.

Where "long" (by my very rough calculations) is going to mean "a substantial fraction of LKO orbital period". An 8-minute burn is a quarter of an orbit... and the Maneuver Planner assumes "instant" dV applied to a specific moment/location. Which in turn adds significant differences between the impulsively-planned maneuver and the actual long-burn-time maneuver. I haven't ever needed to burn longer than about 5 minutes in any of my trips to date, and even that was starting to noticeably adversely affect the actual trajectory vs the planned trajectory. (And Minmus's SOI from 46Mm is a huge target compared with, say, targeting Dres's SOI on the other side of the Kerbolar system.)

If I were doing this with my professional hat on, I'd crank up one of the trajectory-optimization packages, integrate the EOM, and get a burn profile that accounted for the time spent in the burn (and associated orbital V-vector change during that time, etc.). But that's Way Too Much Like Real Work...and I know most of you don't have those sorts of tools ready to hand (and therefore are doing it--successfully--without such tools.)

So I'm casting about for an accessible, low-tech, play-friendly solution. The first thing I thought of was to start by climbing into a much higher orbit with a longer period, so that it's still a reasonable approximation to model a 10+ minute burn as impulsive-dV. But that technique is wasteful of the selfsame fuel that I am using nukes to *conserve* in the first place. And I haven't seen anyone calling that out as a convenient way to improve the accuracy of interplanetary departure burns, so I assume you folk have hit on something else.

So how do you more-experienced Kerbonauts approach the problem of maneuver planning for interplanetary departures with very long burn times?

[Dangerous_Beans]

from what i have read, the usual approach is to split the burn over two orbits; burn to establish a high apoapsis eccentric orbit, then do the injection burn at the periapsis of the next orbit. you then probably need to do a correction burn midway there.

my approach is to just leave from a high orbit and just use a bit more fuel. however i use mods like Kethane or Interstellar, so if i am doing things with low T/W nuke craft i usually have refuelling stations on Minimus so i don't really care.

[Jimbimbibble]

A good rule of thumb is to start burning half of your burn time before the node. It's not perfect but it usually gets you close enough. You will almost always have to make course corrections as it is no it's not such a big deal. A mod that could really help is protractor, which gives you information about phase angles and when to start you burn. Sometimes it works, sometimes it doesn't.

[Claw]

I've heard several people use 4 minutes as a cutoff for splitting up the burn into multiple segments. If it'll take more than 4 minutes, split the burn into two parts. For each burn, split it in half around the maneuver node like you would for a shorter burn. Start two minutes before and stop two minutes after. Place a new node and do the rest of the burn.

[Seyvern]

The best answer is Dangerous_Beans, not Jimbims.

Basically - the idea is - as you point out - you want to burn low where the oberth effect is maximum. Because the orbital period in LKO is very fast (~30 minutes) relative to the inerestion point for a planetary transfer (months to years), you can spread the burn over quite a few Kerbin orbits and not impact your burn trajectory in any noticeable way.

So, step 1, make a maneuver plan.

Step 2, Burn for some short amount of time around hte burn point, this will make your orbit eliptical.

Step 3: Replan your manuver point for the next orbit. Orbit around back to periapsis and burn again. Orbit becomes more eliptical.

Step N: keep repeating step 3 until you have finished your burn sequence.

Two big things to watch out for. Make sure none of your burns intercept Mun or Minmus. Make sure your next to last burn stays within kerbin's SOE

-Seyv

[Kryxal]

I don't like to add more than about 800 delta-v on the initial burns, that's getting you near to the Mun's orbit ... but at that point you're much better off than you were in terms of burn length. Now your 10 minute burn is more like a 6 minute burn, and you're moving faster besides, which will get you clear of the planet (and the area of strongest gravitational attraction) all the sooner, so you'll probably hold course a BIT better.

[Jouni]

I've done a lot of launches to Jool from a 120 km orbit with initial TWR around 0.2. Just start the burn around 5 minutes before the maneuver node, and be prepared to plan a course correction as soon as the burn is finished. Sometimes you have to manually determine when you have completed the burn, because the game tracks long burns incorrectly on low orbits.

Another option is to start from a bit higher orbit. 600 km is about as efficient as lower orbits, assuming that you refuel the ship before launch, but long burns can be made more accurately from there.

I've never liked very much the idea of splitting a long burn in multiple parts. Too much hassle for marginal delta-v savings.

[Pecan]

^^ This.

All of it, especially the last sentence.

[Streetwind]

How exactly does the game do long burns in low orbits incorrectly? I'm always on the lookout to learn more about how stuff works

[Oafman]

I've never liked very much the idea of splitting a long burn in multiple parts. Too much hassle for marginal delta-v savings.

The savings aren't marginal. Nuclear engines are by far the most efficient interplanetary propulsion, and if you need to move a very large ship, then you're stuck with a very long burn, even if you use loads of nukes.

I recently had a 30 minute burn, which clearly needed to be broken down (into 5 burns). With such a big ship, the only way to have made this quicker was with bigger engines (not moar nukes, i already had 24), which would have meant a very significant loss of efficiency.

There are a couple of things you need to be careful of with such long burns:

Firstly, watch your periapsis if you're moving to a solar orbit outside of Kerbin's (ie your node is just after dark). If you're burning from a low orbit, looking to capitalise on Oberth, and you start burning some time before your periapsis, you'll push it down. The equal burn you make after periapsis pushes it back up, but you just need to make sure that you have room to avoid going into the atmosphere. eg, if you're making a 6 minute burn from a 75k orbit, the 3 minutes you burn before the node will probably push your periapsis into the atmosphere. So either start from a higher orbit, or split the burn.

Secondly, if you're doing a really big burn, split into multiples, then you need to think about starting well in advance of your node. Possibly days in advance. People have talked about getting into an eliptical orbit coming close to that of the Mun, and of getting as much velocity as possible without reaching escape. If you do this, bear in mind how long your orbital period will become. Orbits out there can take days to complete, by which time you may have missed the most efficient transfer window.

[Jouni]

How exactly does the game do long burns in low orbits incorrectly? I'm always on the lookout to learn more about how stuff works

I've never bothered to determine exactly what happens. It has something to do with gravity, and the main effect is that you burn longer than you should, because the counter goes down too slowly towards the end of the burn. In extreme cases, you can't even finish the burn, because the counter starts going up at some point.

[metaphor]

If you want to do a single long burn, keep an eye on your trajectory departing Kerbin. Especially the small circle that shows where you exit the SOI. When you get near the end of your burn, stop relying on the burn counter, and instead just try to match up the actual SOI exit circle with the planned one from the maneuver node. This might take a bit more delta-v than the counter, and might mean you burn a little in a radial direction. If you match up the direction on the two SOI exit circles, and also match the departure time, your resultant solar orbit will be exactly what you planned with the maneuver node.

[NecroBones]

How exactly does the game do long burns in low orbits incorrectly? I'm always on the lookout to learn more about how stuff works

The main thing is that, because you're in a low orbit, you are going to swing through a significant arc while executing your burn. During a good portion of this, if you're following the maneuver's marker on the nav-ball, you're partially burning radially (not precisely on the prograde/retrograde marker). The longer you're off the prograde/retrograde mark, the less efficient your maneuver will be.

The maneuver node assumes you are going to change your velocity instantly at that point, so the longer you spend arcing through the orbit while burning, the less precise the results will be.