keep missing my destination on long interplanetary burns

[1bdI]

Ok first things first, I usually bring lots of extra fuel for interplanetary destinations, so my ships tend to be bigger than they probably should be. I use nuke engines since they are so fuel efficient, but with all the extra weight, I end up with very long burns times, (usually 10+ minutes, and that's just to get to eve, duna, dres, etc).

During these long burn times, my manuever node marker (blue one) starts to drift on the navball. I even tried lining it up dead center via the cockpit view. I assume its correcting the course, so I try to stay centered on it. But by doing this I sometimes end up where my required maneuver thrust value has pretty much stopped decreasing. To correct it, I have to hunt around in the general area on the navball till I find somewhere where it starts decreasing at a decent rate again. Once its done, I'll end up close to where I want to be, but it requires a second burn to correct the first burn, wasting even more fuel and time.

Is there something I am missing here? Should I not be following the maneuver marker?

Thanks

[WafflesofMassDeliciousnes]

One of the problems with the interplanetary transfers is that your initial escape burn towards the destination typically doesn't put you close enough to be captured by it's gravity. You will have to course correct once you're about 1/2 to 3/4 of the way there (possibly even both). It isn't a lot, but it is the difference between having a closest approach of 500km and 50km. Your main focus will be the periapsis of your craft as that "should" be the approximate closest approach with your destination. You want that to practically smack into the other planet's surface. If done well you should end up with a slingshot effect that you can burn into an orbit or a landing.

Useful sites I can point you to would be these:

Basic explanation and some pictures

http://ksp.olex.biz/

More detailed explanation and pictures using Jool as the destination.

http://forum.kerbalspaceprogram.com/threads/33547-Tutorial-Non-Hohmann-Interplanetary-Transfers

Launch window planner.

http://alexmoon.github.io/ksp/

A Reddit chain about this exact problem from other KSP members.

http://www.reddit.com/r/KerbalSpaceProgram/comments/1ebce0/how_do_you_do_precise_planetary_transfers/

[Oggula]

In order to best follow the maneuver node course, you should begin your maneuver to carry out half of the delta-v change before the maneuver, and half after. For example a 10 minute burn should start 5 minutes before and finish 5 minutes after. If you are not doing this (correct me if I'm wrong) then most of your Delta-v change will be taking place in a position far removed from the maneuver node.

It is also likely that your long burn times have the same effect; you are not burning where the maneuver node actually is. Try using more nuke engines; I usually assign 1 to every 1600l of fuel. This way, you will also achieve a fuel saving from the Oberth effect, as most of your burn will occur deeper in Kerbin's gravity well.

[Oggula]

To clarify, the blue marker starts to drift as you have moved so far in your orbit, your destination has also moved significantly on the horizon relative to the direction of your orbit, so you must change your direction of thrust to compensate. When the required maneuver thrust stops dropping, that means you are so far removed from the original maneuver course that it is now unachievable.

In my opinion you should clearly be achieving your maneuvers in less time, but if you don't want to add more engines, try executing the maneuver over 2 or three orbits like Scott Manley does here

at around 5 minutes.

[Moonfrog]

I do this for long burns. Heres an example of what I do. Burn time is 2 minutes. Time till node is 1 minute. I begin my burn. This way I average out the long burn around the point of the node. I find this works for me.

Whenever I start at 0 on long burns I always overshoot. So I just half the burn time and start early at that time.

Usually puts me close enough that a little burn forwards or backwards usually enough to get me dead on or fiddle with the apopashis bah I can't spell thing for entry into orbit.

[Kasuha]

I do this for long burns. Heres an example of what I do. Burn time is 2 minutes. Time till node is 1 minute. I begin my burn. This way I average out the long burn around the point of the node. I find this works for me.

The problem is with long burns. If the burn is 15-30 minutes long, you may not be able to start it long enough before the maneuver because you'd lower your periapsis too much and hit the atmosphere. And if you start the last moment when you can, you still get significant difference at the end of the burn compared to what you planned due to various effects.

So far it seems to me when you start diverging significantly from the planned burn it's better to delete the maneuver and prepare a new one.

Also if the ship is really big and heavy, adding a few more nuclear engines won't decrease its dv too much and may improve TWR significantly.

[BlazeFallow]

As others have mentioned, spliting your burn so that you are half-way through the burn at T-0 seconds is the best method for burns.

As an addition, someone did some science on it for KSP .22, and it may have been in effect for 0.21 as well but who knows: when transferring between SOI's it turns out it is best to do so at x5 warp. The exact reasoning is hidden somewhere in the code, but basically at normal speed (i.e. x1 or no warp) the physics engine is still plugging away minute calculations and somehow it glitches when switching SOI's. The same thing happens at higher warps. However, if you set yourself up for a 500km approach to a planet while still in Kerbin's SOI and then do a x5 transfer to Kerbol's SOI and then subsequently to the planet's SOI, you'll end up WAY crazy closer to that 500km mark than you would going through the SOI window at any other warp.

[BlazeFallow]

The problem is with long burns. If the burn is 15-30 minutes long, you may not be able to start it long enough before the maneuver because you'd lower your periapsis too much and hit the atmosphere.

My goodness, burns of 15-30 minutes while you're still in a low orbit around a planet would indeed be suicide. In that case it is best to first get a proper escape burn giving you a suitable escape trajectory for your final maneauver. Make the burn to get on that trajectory and then once settled in that create another node later down the curve to fine tune your path for interplanetary capture.

Keep in mind that interplanetary transfer windows can actually be pretty huge, so even if you wait until you are halfway tot he moon on your escape path you aren't going to be wasting that much dV due to a later transfer burn. A few hours in either direction aren't going to kill you

[wasmic]

My advice: make shorter burns. My Duna burns are usually around one minute for one-way missions, achieved by simply using the LV-909 engine instead of heavy nervas. For very small interplanetary vessels, an LV-909 might actually provide more delta-v than the LV-N. However, as your mission seems to be quite big, I'd say 1) make a booster for it, made with LV-T30 engines, which decouple once out of fuel, and/or 2) add more LV-N's.

[EtherDragon]

Getting to another planet is a series of burns in the right places to minimize fuel used to get an encounter. Using a ship with a low Thrust to Mass ratio makes it more difficult, but it's not impossible by any stretch. Here is a list of maneuvers to plan and execute.

1. Escape Kerbin's Sphere of Influence (SOI)

Based on your description, it may be that you need to execute your Kerbin Escape burn in several phases. If it takes you 10 minutes to complete a burn, by the time you have finished, you are no longer adding velocity with respect to your orbit. Instead, try doing several shorter burns (all at the Periapsis) of no more than 2 minutes each. You only need to repeat these burns for as long as it takes to set your coarse to leave Kerbin's SOI.

2. As soon as you leave Kerbin's SOI - plan and execute a course correction to get an encounter with your destination planet. Since this is in interplanetary space, the length of the burn is no longer a factor, like when attempting to escape Kerbin.

3. When you are at an Ascending or Descending Node, plan and execute another coarse correction to trim your destination approach.

4. As soon as you reach your destination planet's SOI, plan and execute another burn to set your periapsis where you want it. Since your ship has a low Thrust to Mass ratio, you will want to practice the art of Aerocapture. Aerocapture is a technique where you dip your Hyperbolic fly-by into the atmosphere, allowing atmospheric drag to slow you down enough to establish an orbit.

5. Adjust your orbit at the Apoapsis, to achieve whatever kind of orbit you want to explore the area. (If I'm doing moon flyby's like in the Jool system, I might have a highly eccentric orbit...)

Quick question, did you build your interplanetary craft to be modular? If so, send up a more powerful engine cluster...

Edited October 25, 2013 by EtherDragon
Corrections, added links

[1bdI]

Thanks for the replies. Starting the burn halfway towards the time helped a lot. I got spoiled doing all those easy science runs to the Mun I guess. In the end, I didn't miss Duna off the initial burn, and only needed to make some minor adjustments to get in closer.

I usually have lots of fuel left over from the atmosphere stage (like 1/3 to 1/2 on average) with a skipper on it to help out with burn times, but it calculates the time based off that engine. Its hard to figure out the correct time when you have to switch engines mid burn.

I tried putting more nervas on using the 2,3, and 4 way adapter pieces, but then I couldn't add a decoupler, so I figured it couldn't be done, unless there is a different way to do it.

[1bdI]

Getting to another planet is a series of burns in the right places to minimize fuel used to get an encounter. Using a ship with a low Thrust to Mass ratio makes it more difficult, but it's not impossible by any stretch. Here is a list of maneuvers to plan and execute.

1. Escape Kerbin's Sphere of Influence (SOI)

Based on your description, it may be that you need to execute your Kerbin Escape burn in several phases. If it takes you 10 minutes to complete a burn, by the time you have finished, you are no longer adding velocity with respect to your orbit. Instead, try doing several shorter burns (all at the Periapsis) of no more than 2 minutes each. You only need to repeat these burns for as long as it takes to set your coarse to leave Kerbin's SOI.

2. As soon as you leave Kerbin's SOI - plan and execute a course correction to get an encounter with your destination planet. Since this is in interplanetary space, the length of the burn is no longer a factor, like when attempting to escape Kerbin.

3. When you are at an Ascending or Descending Node, plan and execute another coarse correction to trim your destination approach.

4. As soon as you reach your destination planet's SOI, plan and execute another burn to set your periapsis where you want it. Since your ship has a low Thrust to Mass ratio, you will want to practice the art of Aerocapture. Aerocapture is a technique where you dip your Hyperbolic fly-by into the atmosphere, allowing atmospheric drag to slow you down enough to establish an orbit.

5. Adjust your orbit at the Apoapsis, to achieve whatever kind of orbit you want to explore the area. (If I'm doing moon flyby's like in the Jool system, I might have a highly eccentric orbit...)

Quick question, did you build your interplanetary craft to be modular? If so, send up a more powerful engine cluster...

I usually just get into a parking orbit, then find an intercept using the maneuver tool and head off. I have not tried leaving Kerbins SOI then plotting. Maybe for my next mission to Eve I will try it.

Also I have used obex site before, sometimes I use it, other times I just go for it, if the delta v cost is reasonable.

I've done several interplanetary probes to Duna and Eve in .21 where I didnt care about the fuel costs so much since it was never coming back. In .22 I only did one probe of Duna, and two manned interplanetary missions. One was a manned landing to Dres, and a flyby of Eve.

I usually aerobrake on the way back to kerbin from the Mun (or just crash into it because its fun). For this mission to Duna I will probably have to Aerobrake, but I haven't done it outside of Kerbin before. It will take a couple of attempts I think.

Also the ship is modular. I did separate ones more in .21 where I would send up the lander, then send up an engine + fuel to drive it, then dock the two. In .22 I have been just sending the whole thing up in one shot to save some time though. (and building big giant rockets is fun).

[Mesons]

To clarify, the blue marker starts to drift as you have moved so far in your orbit, your destination has also moved significantly on the horizon relative to the direction of your orbit, so you must change your direction of thrust to compensate.

When you drop a maneuver node, once you have configured the burn your are simply given a vector: one direction to burn and the amount of delta-v you need at the exact point you have put the maneuver. If you thrust in the marked direction your orbital position and velocity won't affect the marker at all.

Based on personal experience (I like to send big huge gigantic interplanetary wobbly rockets with too few nuclear engines too), I think I might know the cause of the problem you were having with the delta-v bar not decreasing. If your rocket has any bending or is being controlled from any part/location that is not facing forward, the nav-ball dot will not point in the direction your rockets will be pushing. SAS units on docked parts can exacerbate bending.

I've had several interplanetary ships being pushed by a booster with an orange tank and regular Clamp-O-Tron where the engines on the back of the orange tank swayed back and forth gradually. If this is happening you'll need to keep an eye on the way your rockets are pointing during your burn--not only does swaying make fine-tuning your maneuver nigh on impossible, it is also wasteful during the burn because you are adding delta-v in different directions that really adds up on a long burn.

[Lheim]

Thanks for the replies. Starting the burn halfway towards the time helped a lot. I got spoiled doing all those easy science runs to the Mun I guess. In the end, I didn't miss Duna off the initial burn, and only needed to make some minor adjustments to get in closer.

I usually have lots of fuel left over from the atmosphere stage (like 1/3 to 1/2 on average) with a skipper on it to help out with burn times, but it calculates the time based off that engine. Its hard to figure out the correct time when you have to switch engines mid burn.

I tried putting more nervas on using the 2,3, and 4 way adapter pieces, but then I couldn't add a decoupler, so I figured it couldn't be done, unless there is a different way to do it.

Try to launch it as the centerpiece, with other tanks/engines attached via the truss-like radial decouplers - six or eight of them, with fuel feeding either directly or asparagus-style to the central tank.

[Kasuha]

Today I tried to make a single-burn mission to Moho from 100 km circular Kerbin orbit. The maneuver looked good, some 1600 m/s ending with an intercept, 12 minute burn with four jumbo tanks and six nervas. I started the burn six minutes ahead of the node. After finishing the burn to within 1m/s I found myself some 5° off planned course, no intercept in sight and to correct it to the right trajectory I needed to perform another ~250 m/s correction burn.

I wish it was possible to pause the game while I am setting up a maneuver. Fixing a faulty maneuver while the ship is at high speed in low orbit is pretty much impossible.