Delta-V planning for Duna Orbit transfers.

[DStaal]

I'm in the midst of planning my first mission to Duna, and I was hoping for some advice on mission design - or, more specifically, what amounts of d/v I am going to need for a few sections of my planned mission.  (And how to best allocate that, maybe.)

The mission is going to be semi-ambitious: ~6 Kerbals, autonomous mini rover, lander, and return trip.  All of that is fairly straightforward to plan though.     The question I have is about my ScanSat probes.  I'll need a couple to take along, to map out the planet before I actually land, but they of course should be in a polar orbit.  I'll be entering on a near equatorial orbit from the epileptic, and I'll want to keep that so I can eject back out on it for the return trip, so I'll need to transfer the scanners to their orbit.

Thus far, I see several strategies:

• Separate the scanners before entering Duna SOI, and have them enter directly to a polar orbit.  Downsides are the possibility of needing simultaneous burns - everything becomes very time sensitive.
• Transfer directly from equatorial orbit to polar orbit.  Downside of high d/v use.
• Transfer to polar orbit via Ike rendezvous.  More complicated, may not actually save d/v.
• Perform initial scans from equatorial orbit, gradually changing to a polar orbit.  Doesn't actually save d/v, but may save time on finding a landing site.  (As I don't intend to land at the poles.)

So I come here asking for advice, on what more experienced Kerbonoughts have thought.    Importantly for me, I'm not actually sure what the needed d/v for a 90 degree plane change around Duna would be - making it hard to compare the options.  (I do have the d/v map for the KSP, so I know the d/vs for transfers between Duna and Ike and so on, but I'm not sure on plane change.)

I'm running US-LS, so I will be time constrained, so that is a factor to think about - I need to scan, land, get my initial survey done, and get back to my interplanetary craft in time for the next transfer window to Kerbin, or I'm likely to run out of supplies/habitation.  (I haven't worked out exactly how long that window will be, but I have the tools to do that.)

So, thoughts?  Suggestions?

[Plusck]

If you separate the scanners at the half-way mark, you can use relatively minor combinations of prograde/ retrograde and radial in/out to change your date of arrival quite significantly.

If your scanners have a significant amount of dv available to them, you should be able to get them there a few days quicker still with a modest prograde + radial out burn, then a correction closer to intercept to ensure they arrive into polar orbit.

Edited April 10, 2016 by Plusck

[Geschosskopf]

6 hours ago, DStaal said:

So I come here asking for advice, on what more experienced Kerbonoughts have thought.    Importantly for me, I'm not actually sure what the needed d/v for a 90 degree plane change around Duna would be - making it hard to compare the options.  (I do have the d/v map for the KSP, so I know the d/vs for transfers between Duna and Ike and so on, but I'm not sure on plane change.)

The easiest thing from a design and operations standpoint is to send 1 SCANsat as a totally separate ship from the ground up.  Your Kerbals have no business in polar orbit, your SCANsat has no business in an equatorial orbit, and your main ship has no business lugging the SCANsat around.  Just make it so the SCANsat will arrive in Duna's SOI with about 2000m/s in the tanks (which is very conservative but easy to do thanks to its small size) so it can ,move from Duna to Ike or vice versa.  Because the SCANsat is so small and light, it won't take much juice to get it to Duna on a non-optimum window so it can go out ahead of the main mission by a month or 2 and already have everything scanned by the time everybody else arrives.

And the SCANsat has no need to spend much dV to get to polar orbit.  It can do that pretty much for free, and it has no need to go back to equatorial to switch from one body to the other.  It goes like this...

Scanning Duna First:

1. When you do your mid-course trajectory tweak, adjust your Pe so that it's over Duna's north pole or under its south pole, doesn't matter which.  This will put you in a polar orbit once you capture.
2. Aerocapture at Duna so that your Ap is at the altitude you want to scan from, discard the fairing that you've kept until this point, and adjust your Pe to circularize.  Then scan Duna.
3. Once done with Duna, you get to Ike by a "hook shot" above or a "massé" shot below Ike's orbital plane.  You leave when your orbital plane is lined up like 1/8 of Ike's orbital circumference ahead of Ike's current position.
4. You will then arrive vertially at Ike and will be in a polar orbit once you capture there.

Of, if you want to scan Ike first...

1. Keep your Duna Pe in the equatorial plane all the way there.
2. Areocapture at Duna into an elliptical orbit with the Ap out near Ike's orbit.  Discard fairing.
3. If you get an immediate Ike encounter coming off Duna, do a mid-course tweak between Duna and Ike to make that encounter polar.  Otherwise, coast to Ap, raise your Pe out of Duna's air, and wait an orbit or 2 for an Ike endounter or make one happen sooner with a small burn, and adjust it so it's polar.
4. Capture into a polar orbit at Ike, adjust altitude as needed, and scan.
5. When done scanning Ike, wait until your orbital plane is pointed at Duna, then do a "hook" or "massé" shot to transfer there, capture and scan.

[FancyMouse]

As mentioned, adjusting SoI entry time should be just adding prograde/retrograde and/or radial in/out factor during mid-course correction burn. It shouldn't cost too much more than the correction for inclination itself. And even better, there should be just a single mid-course burn that both adjusts SoI entry time as well as changing inclination. This both saves fuel and player time (don't need to launch a new ship and plot a new interplanetary maneuver and worry about yet another correction etc.)

So - you can go with your option #1 with your concern solved.

Edited April 11, 2016 by FancyMouse

[GoSlash27]

DStaal,

If you eject the scanners at the mid-point of the interplanetary transfer, the DV required to get them polar vs. equatorial will be so slight as to be inconsequential.
You can adjust with tiny burns to get a polar intercept, then barely close the orbit either by retroburn or aerobraking.
If you leave a high Ap, corrections to the inclination will be very cheap when you reach Ap because your velocity will be so tiny.

Finally, circularize the polar orbit either by aerobraking or retroburning.

Good luck!

-Slashy

[Nich]

Yes mid point separation.

Aerobrake and bring PE down for Ike encounter. Scan

Wait until polar orbit lines up with pro/retrograde Burn for Duna

Aerobrake Scan

Your biggest dv investments are Ike capture and exit

[Badsector]

Little suggestion is install transfer window planner and you can do a lot of calculations for planning the trip, for cover your question a normal transfer to duna require ~1600 dv from a 100km kerbin orbit to 100km duna orbit and 305 days of flight (~1100 ejection, ~500 retroburn) but with 2000dv you can do the trip in 200 days or 150 days with 2800dv .

Duna is enough easy and with few dv you can speed up a lot the trip

[DStaal]

Thanks all for your answers.  @Badsector, I do already have Transfer Window Planner, and it's very useful.    I was asking about d/v for the plane change transfers, so I could compare them to the other options where I already have the d/v.  (From Transfer Window Planner and the KSP system map.)

However both you and @Geschosskopf make a good point: For a small extra d/v investment, I can launch the scansat satellites ahead of time direct from Kerbin.  I am somewhat experienced with injecting directly to polar orbit - I've been doing that for scansat launches to Mun and Minmus all along.  (And I've been doing more of those satellites than I'm planning for Duna, as I would tend to launch as new scanners became available.  The Duna satellite will be a 3-in-1, for the different needed altitudes, but it will have a full compliment of scanners.)

So, my current plan is to launch a Duna scanner sat ahead of the main ship, using ion drives.  (The main ship will use either chemical or nuke propulsion, to make possible a refuel on Duna if needed.)  The main ship will carry a second 3-in-1 block, intended for Ike, but it can be used as a backup in case something goes wrong with the Duna scanners.  This simplifies the planning, and avoids the mid-transfer separation (which sounds easy enough, but makes me nervous), and gives me the experience with larger ships that I want for future missions to other planets.

[Plusck]

2 hours ago, DStaal said:

So, my current plan is to launch a Duna scanner sat ahead of the main ship, using ion drives.  (The main ship will use either chemical or nuke propulsion, to make possible a refuel on Duna if needed.)  The main ship will carry a second 3-in-1 block, intended for Ike, but it can be used as a backup in case something goes wrong with the Duna scanners.  This simplifies the planning, and avoids the mid-transfer separation (which sounds easy enough, but makes me nervous), and gives me the experience with larger ships that I want for future missions to other planets.

For a given transfer window for Duna, you can't meaningfully "launch ahead". All you can do is be more or less efficient in your transfer.

The most absolutely efficient Hohmann transfer will take exactly half of an orbit, arriving exactly 180° from your starting point, and will only just brush against the orbit of your target.

If you launch "ahead" of time, you're necessarily spending more fuel but you're not necessarily going to get there sooner - that depends on your ejection angle. Since you are starting earlier than you should, you can get there sooner by ejecting with a radial out component (with respect to the sun) but then you'd get exactly the same effect for less cost if you did that during the optimal transfer window...

If you look at the blue porkchops on the alex moon transfer window planner, they're more or less vertical. That means you can launch at the same time but get there sooner or later, for much the same cost, just by altering the ejection angle. For Eve, the smudges slope down to the right, meaning that if you launch sooner you'll need to spend much more fuel or else you'll get there later. For the outer planets it's the other way round, you can launch sooner for Jool and get there sooner, for much the same cost, but the benefits are still minimal since launching at the same time can give you a 40-day difference with just 10-20 m/s difference in ejection burn, or mid-course correction burn.

Edited April 11, 2016 by Plusck

[numerobis]

The way NASA would do it is to launch the support craft in one transfer window, the humans in the next transfer window. That way, if the support craft fail at any point in the trip, you can just scrub the human launch without anyone dying.

[DStaal]

By 'ahead' I mean maybe a day or two, using transfer window planner to find a solution that gets me to Duna ahead of the main ship.  I've got a plot going using that and Maneuver Node Splitter so that I'm not trying something dumb like doing a burn for more than a full orbit.    Basically, it should get there first, from a separate launch.  Hopefully.

And I thought about doing a smaller first mission - just a couple of robotic probes, doing an initial survey - but that would put my long-term plans so far in the future I don't want to deal with it.  This game is slow enough as it is - typically my progress so far is less than real time, so spending thousands of days getting this done doesn't sound like fun.  NASA spends time because they have to, but this is a game and I don't want to spend super-huge amounts of time on it. 

[Plusck]

1 hour ago, DStaal said:

By 'ahead' I mean maybe a day or two, using transfer window planner to find a solution that gets me to Duna ahead of the main ship.  I've got a plot going using that and Maneuver Node Splitter so that I'm not trying something dumb like doing a burn for more than a full orbit.    Basically, it should get there first, from a separate launch.  Hopefully.

Ok - but please believe me that the accuracy and angle of your transfer burn will have a far greater effect on arrival time than sending them a day or two earlier.

Doesn't mean you can't send your most disposable ships first, to test the waters, and deliberately burn slightly outwards to meet Duna sooner.

Another thing to bear in mind is the time that your ships cross the SOI boundary. You don't want to be on very high warp (six or more arrows) when you do that. Since it'll take a day or two to reach the edge of Kerbin's SOI, be very careful with warping while you're controlling the second wave. The reason is the accuracy of the resulting orbit as you change the SOI. You don't want to do this while in control of the ship in real time (a game engine limitation) and you don't want to do it at very high warp (because the game tests for an SOI change at larger and larger intervals).

Edited April 12, 2016 by Plusck