Periapse kicking for rendezvous.

[qoonpooka]

So I've launched my first-for-this-career (second-ever) space station.  I have a contract to park it over Minmus and I would like it to live there anyway because I want to use it as a refueling stop for interplanetary journeys. (I figure I'll rectify the orbit's angle prior to departure.)

The station has very low TWR because it's not intended to move around much. Even with an inter-orbital tug attached it'll still need perikicking to make it to minmus.  

My usual method won't work here. Usually I click on the orbit, add some prograde burn, wait for the conics to go all screwy, make minor tweaks until I like my arrival characteristics, time warp to maneuver node, make the burn.

The problem is, by the time I'm done orbiting a couple of times, Minmus will have gone along on its merry way.

Is there a guide to knowing where in a circular parking orbit you need to begin kicking for a Kerbin->Minmus transfer? I can time it for an AN/DN intercept if that helps, or I can match alignments in my parking orbit first.  

[Hupf]

Related:

 

[qoonpooka]

23 minutes ago, Hupf said:

Related:

 

Indeed.I don't understand that spreadsheet at all, nor why I have to give the progression in terms of apoapsis height rather than acceptable burn time, etc.

I was hoping there was a guide somewhere that explains how to use that tool, especially since I am now realizing I'm about to lose a Duna launch window and want to get a probe out there ASAP.

[Empiro]

I think I understand your issue -- you need to raise your AP slowly over multiple orbits, but it's difficult to meet with Minmus right when you're at AP correct?

It's actually not very difficult, as long as you're not in a time crunch. Simply burn at PE until your AP is touching Minmus' orbit. Make sure that you're on the same inclination as Minmus (or at the least, the AN or DN is at your AP, which is touching Minmus' orbit). Then, assuming your orbital period isn't exactly in some sort of phasing orbit with Minmus' (which is nigh impossible), you'll eventually get an encounter at AP.

If you want to speed things up, what you can do is place a maneuver node at AP, and place another one right after it. Click "next orbit" on the second node until you get a close encounter where your ship is ahead of Minmus. Go back to the first maneuver node and add a burn slightly prograde. You might need to readjust your second node's position, but even a tiny burn should let you get an encounter.

Essentially what you're doing is increasing your orbital period by burning at AP, that small difference quickly adds up over the course of a few orbits, which slows you down and lets you get that Minmus encounter fairly quickly.

[qoonpooka]

1 hour ago, Empiro said:

I think I understand your issue -- you need to raise your AP slowly over multiple orbits, but it's difficult to meet with Minmus right when you're at AP correct?

It's actually not very difficult, as long as you're not in a time crunch. Simply burn at PE until your AP is touching Minmus' orbit. Make sure that you're on the same inclination as Minmus (or at the least, the AN or DN is at your AP, which is touching Minmus' orbit). Then, assuming your orbital period isn't exactly in some sort of phasing orbit with Minmus' (which is nigh impossible), you'll eventually get an encounter at AP.

If you want to speed things up, what you can do is place a maneuver node at AP, and place another one right after it. Click "next orbit" on the second node until you get a close encounter where your ship is ahead of Minmus. Go back to the first maneuver node and add a burn slightly prograde. You might need to readjust your second node's position, but even a tiny burn should let you get an encounter.

Essentially what you're doing is increasing your orbital period by burning at AP, that small difference quickly adds up over the course of a few orbits, which slows you down and lets you get that Minmus encounter fairly quickly.

This seems like a good hack to get the job done for Minmus, but when I'm looking at being efficient with dV near Jool, for example, or looking to transfer these skills to interplanetary transfers, I'm in not-so-good shape again.  I'd like to know how to compute the timing to get it just-in-time so that I can build those skills for my upcoming Eve and Duna runs.

[Plusck]

29 minutes ago, qoonpooka said:

This seems like a good hack to get the job done for Minmus, but when I'm looking at being efficient with dV near Jool, for example, or looking to transfer these skills to interplanetary transfers, I'm in not-so-good shape again.  I'd like to know how to compute the timing to get it just-in-time so that I can build those skills for my upcoming Eve and Duna runs.

Empiro's answer (particularly the third paragraph) is what I do and it is very effective for most precision burns.  For your case, if you are making (say) three separate Pe burns, just add them all in sequentially. It helps to have timewarp on x5 (lowest warp) to stabilise the orbital calculations.

Of course, by the time you get to the third burn, errors will have racked up but they should be relatively simple and painless to correct for the last burn.

[qoonpooka]

1 minute ago, Plusck said:

Empiro's answer (particularly the third paragraph) is what I do and it is very effective for most precision burns.  For your case, if you are making (say) three separate Pe burns, just add them all in sequentially. It helps to have timewarp on x5 (lowest warp) to stabilise the orbital calculations.

Of course, by the time you get to the third burn, errors will have racked up but they should be relatively simple and painless to correct for the last burn.

How do you put maneuver nodes on top of each other?

[Plusck]

19 minutes ago, qoonpooka said:

How do you put maneuver nodes on top of each other?

Simple answer: don't, quite.

Option one: add the second burn node just behind the first one (the game will automatically interpret this to be one orbit later), set it to the right dv number, then ever so carefully reduce the first burn which will drag the second node over Pe.

Option two: as above, but don't bother trying to bring the nodes over each other, instead adding a slight radial-in component and starting the actual burn a touch later. Delta-v savings for a 20s difference in starting position will be minimal, and if you start the burn 20s later you'll automatically include the radial-in component for zero actual cost!

In any event, if you have a very low-TWR craft your cosine losses will be increasingly large on each burn. It's pointless trying to aim precisely for Pe since a a vast majority of the burn will not be anywhere near it.

 

Final, final option: start in a higher orbit. It takes more dv to get there to start with, but escape velocity is lower and cosine losses for low-TWR burns are much lower. IIRC the ideal orbital altitude for going to Jool is about 300km - so while it is less efficient to push LKO up to that height before making the interplanetary burn, the interplanetary burn itself will be shorter at that altitude than in a 100km LKO.

Edited August 26, 2016 by Plusck

[qoonpooka]

1 hour ago, Plusck said:

Simple answer: don't, quite.

Option one: add the second burn node just behind the first one (the game will automatically interpret this to be one orbit later), set it to the right dv number, then ever so carefully reduce the first burn which will drag the second node over Pe.

Option two: as above, but don't bother trying to bring the nodes over each other, instead adding a slight radial-in component and starting the actual burn a touch later. Delta-v savings for a 20s difference in starting position will be minimal, and if you start the burn 20s later you'll automatically include the radial-in component for zero actual cost!

In any event, if you have a very low-TWR craft your cosine losses will be increasingly large on each burn. It's pointless trying to aim precisely for Pe since a a vast majority of the burn will not be anywhere near it.

 

Final, final option: start in a higher orbit. It takes more dv to get there to start with, but escape velocity is lower and cosine losses for low-TWR burns are much lower. IIRC the ideal orbital altitude for going to Jool is about 300km - so while it is less efficient to push LKO up to that height before making the interplanetary burn, the interplanetary burn itself will be shorter at that altitude than in a 100km LKO.

What do I do with all this stuff about phase angles then? Just bin it?

[Plusck]

38 minutes ago, qoonpooka said:

What do I do with all this stuff about phase angles then? Just bin it?

No, but it's impossible not to compromise unless every maneuvre is done with a precision that you simply cannot plan with stock KSP.

Stock KSP maneuvre nodes assume instantaneous changes in velocity. In LKO, any burn that lasts for more than a few seconds will incur slight losses. If it lasts for a couple of minutes, you will not leave on the exact path that was planned because each second of burn will carry the "central point" of your burn further and further away from the point that the maneuvre node was originally placed.

If you are going interplanetary from LKO with a low-TWR craft, that deviation from the planned flight-path will be significant, unless you plan for it. If you plan for it, you have to accept the correction dv losses in the final phase of your burn. If you are going to accept correction losses, you might as well plan ahead by including a minor radial component on slightly offset maneuvre nodes then manually changing the burn starting time to eliminate the implied loss.*

And if you are going interplanetary with a very low-TWR craft, you're making it hard for yourself starting in LKO. A higher orbit would allow longer burns and les deviation, but at a dv cost of getting to that orbit to start with. Again, a compromise.

So there's no need to forego accuracy; you just need to decide what to compromise on, which is probably going to be dv losses.
If the only thing you want to compomise on is the time it takes to set it all up, then you should really be placing dozens of maneuvre nodes each representing only a couple of minutes of burn time. Having done this just the other day to capture at Dres with an underpowered ship (16 minute burn!) that only just had enough fuel to capture (about 150m/s more than an instantaneous burn at Pe over Dres would have needed), I can confirm that it works, but it's a pain to set up and it gets utterly messed up if you overrun a node. In the end, the planning merely told me when to start burning directly retrograde and what angle I needed to be approaching at to maximise available dv...

 

* For example, if you place a maneuvre node and set it to achieve a precise destination such as a difficult planetary encounter, then move the node back 30s along the orbit and add radial-in and subtract a very small amount of prograde to correct until you achieve the exact same encounter... now obviously if you start that second burn 30s later than you would have anyway, you will be pointing exactly the same direction and you will burn exactly the same amount of fuel as if you had just used the initial maneuvre node.

 

 

[qoonpooka]

2 hours ago, Plusck said:

No, but it's impossible not to compromise unless every maneuvre is done with a precision that you simply cannot plan with stock KSP.

Stock KSP maneuvre nodes assume instantaneous changes in velocity. In LKO, any burn that lasts for more than a few seconds will incur slight losses. If it lasts for a couple of minutes, you will not leave on the exact path that was planned because each second of burn will carry the "central point" of your burn further and further away from the point that the maneuvre node was originally placed.

If you are going interplanetary from LKO with a low-TWR craft, that deviation from the planned flight-path will be significant, unless you plan for it. If you plan for it, you have to accept the correction dv losses in the final phase of your burn. If you are going to accept correction losses, you might as well plan ahead by including a minor radial component on slightly offset maneuvre nodes then manually changing the burn starting time to eliminate the implied loss.*

And if you are going interplanetary with a very low-TWR craft, you're making it hard for yourself starting in LKO. A higher orbit would allow longer burns and les deviation, but at a dv cost of getting to that orbit to start with. Again, a compromise.

So there's no need to forego accuracy; you just need to decide what to compromise on, which is probably going to be dv losses.
If the only thing you want to compomise on is the time it takes to set it all up, then you should really be placing dozens of maneuvre nodes each representing only a couple of minutes of burn time. Having done this just the other day to capture at Dres with an underpowered ship (16 minute burn!) that only just had enough fuel to capture (about 150m/s more than an instantaneous burn at Pe over Dres would have needed), I can confirm that it works, but it's a pain to set up and it gets utterly messed up if you overrun a node. In the end, the planning merely told me when to start burning directly retrograde and what angle I needed to be approaching at to maximise available dv...

 

* For example, if you place a maneuvre node and set it to achieve a precise destination such as a difficult planetary encounter, then move the node back 30s along the orbit and add radial-in and subtract a very small amount of prograde to correct until you achieve the exact same encounter... now obviously if you start that second burn 30s later than you would have anyway, you will be pointing exactly the same direction and you will burn exactly the same amount of fuel as if you had just used the initial maneuvre node.

 

 

Your Dres example is basically what I'm aiming for here. I'm using equipment that is probably not suited to capture to Duna, but my window closes in 2 game-days so if I'm doing Duna anytime this game-year, I need to launch it NOW.  But the stuff I have means it's going to be a 0.5 TWR probe (because of all the fuel it needs to haul for the dV).  Most of our conversation has gone completely over my head, so I guess the answer to my OP is: No. Such a guide does not exist.

[Plusck]

Capturing at Duna isn't too bad. You can help it along by passing in front of Ike first for a gravity assist (Ike deflects you into a more vertical path towards Duna at a slightly lower orbital velocity, making capture at Duna a touch cheaper). You can also burn retrograde for a while at Ike for a more pronounced effect - but it's still cheaper to do most of the capture burn close to Duna. What I mean is this:

Spoiler

Apologies for the 1.0.5 graphics:

The burn at Ike here is minimal - merely a correction since I am wanting to match orbits with that ship arked in blue.. By using combinations of radial/prograde during a half-way correction burn, you can arrange it so that you meet Ike a touch later in its orbit, pass in front and burn retrograde at Ike Pe to drop closer to Duna. It helps to reduce the final Duna burn if you don't want to aerobrake. Total capture dv here is about 780m/s, but that's mostly because I'm going straight into the target orbit rather than capturing at a 60km Pe like I usually would.

And a TWR of 0.5 is not "very low"