Duna Spacecraft Help -- Career

[saamoz1014]

So I started playing just one update ago (0.23 for everyone who's been under a rock), and I finally decided to go to Duna. I know I've only started playing, but I like to think that I am not *that* bad, if inexperienced. In career mode, (which I'm still working on), I get whatever mods look interesting, which currently include Mechjeb (I refuse to use the automatic transfer. Period.) I also have remote tech (which complicates unmanned missions). On the tech tree, I don't have atomic engines, and I have the very basic rockomax parts (not the big capsules or anything though.) Oh, and I just got the first clamp-o-tron like an hour ago. My problem is that I wanna go to Duna, but I want to return too. I barely have enough fuel for a flyby, and I can't get my periapsis low enough to aerobrake; it's requires too much precision. However, I think the main problem is my ship design. I heard somewhere that you need ~7500. I have that amount, but I think that includes aerobraking and that pro veteran stuff that's beyond me.

Picture of my tech tree:

My Duna Rocket:

Again on the Launchpad

The first stage is the little orange solid boosters and 7 big long white ones. After that, I have 7 F-T800s in asparagus staging, all with LV-T45s. This gets me into a nice parking orbit at around 90km. At the launch window (44.36 degrees) I use my next stage; 3 LV-909s on 3 F-T800's. When I get to Duna, I hve a little less than half left. If I land, then the lander with two F-T400s and LV-909s can't get back home, so I have to go straight back. Please help, as I want more science so I can start building space stations and planetary bases.

[Hodo]

I would first build a station around Kerbin, just to refuel your Duna ship. It would make life considerably easier. And the Rockomax parts are almost a must, but it can be done without them, it is just alot harder.

I myself have only been to Duna once with a space plane SSTO I built before .23. It was the last thing I did in .22 and it only managed to land on Ike because my orbit of Duna was unfavorable for landing where I needed to, to refuel the ship.

[Taki117]

I suggest this for aerobraking help. This to save some dV getting there and back. and This for figuring out how much dv you need to get there and back.

[SpaceGremlin]

Here's the mission I sent out to Duna a few days ago.

I think the only techs I had that you don't are Heavier Rocketry, Electronics, and Advanced Science (I rushed to the end of the science branch to maximize my interplanetary trips, but its optional).

I assembled it in LKO with several launches (much easier), and the drive section is a KW SPS (similar to a Poodle engine, just a touch better). Its meant to clean the entire Duna system out of science in one go (high/low Duna/Ike orbits, Duna and Ike landings, etc). All the science is piled into the 3 man pod to be returned to Kerbin with the crew.

As for getting the intercept, here's what I do: when my transfer window opens up I put my apoapsis out to Duna orbit and play with the node until I get an intercept with Duna's SOI. After performing that burn I trim the intercept down to low Duna orbit (I usually do this at the ascending/descending node). Simply adjust normal/antinormal to get the intercept to the desired inclination, and then its usually a slight prograde/retrograde or radial/antiradial adjustment to get the periapsis down below Ike's orbit. The further away you are, the less dV you'll need to adjust, but the less control you'll have. So don't be afraid to do this multiple times as you get closer. I usually set my final aerobrake altitude by burning radial/antiradial at the edge on Duna's SOI.

[Jouni]

If you use (almost) optimal transfer windows, a return mission to Duna requires only a little more fuel than a similar mission to Mun. The mission profile is a bit different, however.

For a mission to Mun, you need 7500 m/s of delta-v, allowing some margin of error. After reaching a parking orbit around Kerbin, you should have at least 2800 m/s left to complete the mission. To be able to return to Kerbin, the lander should have at least 1000 m/s remaining at takeoff.

In a Duna mission, the total delta-v requirement is around 8000 m/s, with a similar margin of error. This assumes either aerobraking at Duna or a using a risky strategy, where you just set your periapsis below 10 km and hope to hit reasonably level ground. Add 500 m/s for establishing a low orbit around Duna before landing, if you don't like aerobraking. At the parking orbit around Kerbin, you should have 3300 m/s remaining (or 3800 m/s, if you don't plan to aerobrake). Reaching Duna and landing at it requires surprisingly little delta-v, as you can use atmospheric drag and parachutes to kill most of the speed. Bringing enough parachutes for a safe landing is not advisable, however. It's more fuel efficient to carry fewer parachutes and do a short burn just before landing to bring the speed to a safe level. The lander should have at least 2000 m/s remaining to return to Kerbin. This also assumes aerobraking or just diving deep into Kerbin's atmosphere - establishing an orbit around Kerbin can be expensive after interplanetary missions.

[Kryxal]

Also, for those fiddly little adjustments when you're half an interstellar orbit away, try using single taps of RCS, does wonders (but plot a maneuver node first so you know what it will do!)

[Jouni]

I kind of enjoy building early career mode ships, but the low-tech phase lasts so little and getting science is so grindy, so I play sandbox instead.

Anyway, I would probably build something like this for an early career mode mission to Duna. The lifter is a bit underpowered, but good enough for the purpose. The rocket takes off with six small solid boosters. After them comes the main lifter stage with three large solid boosters (set to 70%) and three LV-T45 engines. Final orbital insertion is done with payload engines.

The mission stage falls a little short of my 3800 m/s target for a non-aerobraking Duna approach, but it can still probably achieve it. The large fuel tank has enough fuel to reach Duna, as well as for possible course corrections on the way. After ditching the empty fuel tank, the lander has almost 2600 m/s remaining for slowing down to orbital speeds, landing, and returning home. The big parachutes on top of the fuel tanks are intended for Duna landing. If you repack them, you can probably bring the entire lander back in one piece. It's still better to transfer the experiments to the command pod, so that you can separate it and land with its parachute, if something goes wrong.

[Shna_na]

I've skimmed the replies but read the OP, and I think you need this. To save you the effort of clicking, it says that Duna requires 5980m/s Delta V, including everything, to achieve an orbit above Duna that's 10Km above the highest point of its atmosphere. Let's assume that, including errors throughout the mission, you need another 1220m/s (if you use parachutes) to land and make up for errors. You then need 1500m/s, including errors for the rest of the mission, to achieve orbit again. All of this adds up to around 8700m/s. Add the original number, minus 4500 of the 4550 needed to reach orbit from Kerbin's surface, and you have 8700+1430=10130. But we'll be safe and call it 1500, so it adds up to 10200m/s.

That's your approximate total: 10200m/s Delta V for the whole mission, including a lot of errors. You'd probably make it with just 10000m/s, but either way it's a lot. Use Kerbal Engineer, it'll let you know how much you have per stage and from there you can do simple addition to see if you have enough.

[Jouni]

I've skimmed the replies but read the OP, and I think you need this. To save you the effort of clicking, it says that Duna requires 5980m/s Delta V, including everything, to achieve an orbit above Duna that's 10Km above the highest point of its atmosphere. Let's assume that, including errors throughout the mission, you need another 1220m/s (if you use parachutes) to land and make up for errors. You then need 1500m/s, including errors for the rest of the mission, to achieve orbit again. All of this adds up to around 8700m/s. Add the original number, minus 4500 of the 4550 needed to reach orbit from Kerbin's surface, and you have 8700+1430=10130. But we'll be safe and call it 1500, so it adds up to 10200m/s.

Let's look at the delta-v map those estimates are based on.

For a minimal Duna mission, we need 4550 m/s to reach LKO, 950 m/s to escape Kerbin, and 110 m/s for Duna intercept. Assuming that we have been accurate enough, the periapsis is deep enough in Duna's atmosphere that we can aerobrake and land without burning any more fuel. In the best case, we need just 5610 m/s to land on Duna. Assuming that we want to return to Kerbin, we need 1380 m/s to return to orbit, 370 m/s for Duna escape, and 110 m/s for Kerbin intercept. Some ridiculous aerobraking brings us back safely, so we have just completed a Duna return mission for 7470 m/s. Add something for course corrections and a margin of error, and we reach my earlier 8000 m/s estimate.

If we don't want to aerobrake at Duna, we need a minimum of 370 m/s to get to low Duna orbit. My earlier 500 m/s estimate was based on this and some additional margin of error, which brings my total to 8500 m/s.

If we don't want to aerobrake at Kerbin either, we need an extra 950 m/s to return to LKO before landing. Let's say this is 1000 m/s, because I like round numbers. My total is now 9500 m/s. The 700 m/s difference between our estimate comes from the part, where you add 1220 m/s for landing and errors.

My example ship can probably complete the 8500 m/s Duna return mission. The 9500 m/s mission requires a different approach, because the lander must have 3000 m/s remaining after landing. While you can certainly build such lander and get it to Duna with a reasonable payload, using only low-tech parts, carrying so much fuel to the surface is not very efficient. An Apollo-style mission with a command module and a lander could be one approach, but it might be easier just to leave an extra fuel tank in the orbit.

[saamoz1014]

I would first build a station around Kerbin, just to refuel your Duna ship. It would make life considerably easier. And the Rockomax parts are almost a must, but it can be done without them, it is just alot harder.

I myself have only been to Duna once with a space plane SSTO I built before .23. It was the last thing I did in .22 and it only managed to land on Ike because my orbit of Duna was unfavorable for landing where I needed to, to refuel the ship.

I'm a little hesitant about building a ship, or refuelling it in orbit, since I only have the basic clamp-o-trons. If I build something, I doubt that those ports can hold it together. By the time I get into orbit, I have three full FL-T800s, so if I want to refuel, I need to add some extra tanks somewhere, which might hurt the aerodynamic profile of the ship (ya, right). But what I'm really having trouble with is having enough fuel for the lander to get home. I need a lander/ship design that will be able to get me home.

[KerbMav]

I can't get my periapsis low enough to aerobrake; it's requires too much precision. ... but I think that includes aerobraking and that pro veteran stuff that's beyond me.

No, it is not, neither beyond you nor pro stuff.

Just keep in mind that the farther away from your target you burn, the bigger the effect will be. Get your encounter first, adjust it half way of the voyage, get it like you want it to be somewhere close to the planets sphere of influence.

Even a bit assistance from the atmosphere can reduce the needed dV to achieve orbit. As long as you get captured by Duna you can pass through the atmosphere a few times to lower your apoapsis step by step.

Either use the linked calculator above or use quicksave and experiment away, learning by doing!

If I land, then the lander with two F-T400s and LV-909s can't get back home, so I have to go straight back. Please help, as I want more science so I can start building space stations and planetary bases.

I need a lander/ship design that will be able to get me home.

Whenever you are leaving something behind you are saving fuel in the next step of your voyage. How good are you with docking/orbital rendesvouzs?

Leave your transfer stage behind and only land the lander with enough fuel to deorbit, brake above the surface if needed, launch back up and dock with the transfer stage again.

You could even leave parts of the lander behind for the return trip - or not bring it back at all and use a second pod on the transfer stage with its own parachute for a direct landing at Kerbin. (Which would add mass to the ship from the beginning though.)

You could definitely leave lots of stuff behind on Dunas surface, like the big science parts (Do not forget to transfer the precious experiments to your pod! ) or the tank used for landing and the landing gear of course.

I only have the basic clamp-o-trons. If I build something, I doubt that those ports can hold it together.

They will do fine for keeping your pod on the transfer stage. For the critical phase of launching the ship you can use struts to stabilize them more. When undocking e.g. the lander at Duna they will vanish for good, but the clamps will surely hold on your return trip, as you will land at Kerbin with the pod only in most cases.

Edited February 17, 2014 by KerbMav

[Jouni]

But what I'm really having trouble with is having enough fuel for the lander to get home. I need a lander/ship design that will be able to get me home.

You can probably achieve that without significant changes to your design by making the lander lighter.

- Small ships usually don't need RCS, except for docking. This saves you at least 0.75 tonnes (0.55 for the fuel tank and 0.20 for the four thruster blocks).

- You may need the reaction wheels when the transfer stage is still attached, but the command pod should provide enough torque for the lander. By moving the reaction wheels to the other side of the decoupler, you save 0.3 tonnes.

- With the RCS fuel tank and the reaction wheels gone, you can move the center of mass of the lander a bit lower. This makes the lander more stable, allowing you to use 4 landing struts instead of 6. Another 0.1 tonnes saved.

- You can use simple solar panels instead of the heavy deployable panels, unless you are using some mod that requires a lot of power. This saves 0.08 tonnes for the four panels.

- There were some ladders attached to the parts you have already removed, and some of those ladders had matching pairs on the other side of the ship. By removing four ladder components, you save 0.02 tonnes.

After these modifications, your lander is 1.25 tonnes lighter. If that does not increase delta-v enough, you can add more fuel to the lander. Two FL-T100 fuel tanks add 1.125 tonnes to the mass, and the lander would still be lighter than originally. FL-T200 fuel tanks could also be acceptable, as your transfer stage has a lot of extra fuel. You could also use half-empty FL-T800 fuel tanks instead of the full FL-T400 tanks in your lander, and fill them from the transfer stage before detaching it.

[radonek]

I'm a little hesitant about building a ship, or refuelling it in orbit, since I only have the basic clamp-o-trons. If I build something, I doubt that those ports can hold it together.

It definitely can. And I dare to say that perfecting such design is way easier then fussing around with direct ascent. You just moar-strut it like any other part as necessary to withstand launch forces. Once on orbit, clamp-o-tron should be strong enough for anything you push him with. You can make it fail by kicking it with some high-TWR beast (yes mainsail, I'm looking at you) but using such ineffective engine at this point is error in itself. Plus, docking equipment can be invaluable if something goes south and you need to pull off a rescue mission.

[Clockwork_werewolf]

With the new tweekables you can send up a larger craft, empty of fuel into Kerbal orbit. Refuel it up there with one or tow other craft and you can have a craft with three times the fuel and engines than you could send up in one launch.

Drag is calculated on mass not surface area (unless you have FAR installed) so don't worry about a large empty final stage.

Its worth noting you can fill the tech tree with just Mun and Minmus science but I understand you wanting to go to Duna

[ArmchairGravy]

This is onion-staged Skippers, X200-32's (Jumbo 64's pictured, but not needed), and an LV-N. Round trip ticket to Duna and land on Ike while you're there.

Get any intercept you can, then when you're at AN/DN work the 6 main directions (ignore markers--burn 0,90,180,270, in, out) to get your periapsis down without losing your intercept. If you need to, re-adjust again when you're 3/4 of the way there. You should be able to get periapsis down to 100-200k. Once you hit SOI you can fine-tune down to 12.5k for aerocapture. Like everything else in KSP it will take some practice, but once you get it, it's got. Good luck!

Edited February 17, 2014 by ArmchairGravy

[KerikBalm]

I recently did my first stock mission to Duna (previous times, I was using self-modded parts to "simulate" a spacecraft with a 1 GW reactor -> 30 kn 7,000 ISP engines for interplanetary travel, 240 kn thermal engine with an 875 ISP, turbines that worked without oxygen - I did not wait for launch windows, and used extreme amounts of delta V, and very very deep aerobraking).

I did it in sandbox (my career mod I'm sciencing as fast as I can at the Mun and Minmus for a Moho launch at day 6, since that is the first interplanetary launch window.)

I then did it again slightly differently.

I *hope* you are waiting for the launch window (first one is about days 57-62 with 59 being the optimal) - that helps a lot.

Then go into your config file, and set your conics mode from 3 to 0. This helps a lot! Once you get your intercept, focus on duna, your trajectory (within duna's SOI) wil display there, and you can actually look at it well. Use RCS (or in my case, I used Ion engines) to set up your orbit to be prograde and equatorial, if you're coming in on a hohman trasfer from Kerbin, you'll want to set your perapsis at 12,000 to 12,000 - Remember not to time warp across SOI changes (ie during kerbin escape, and when entering Duna's SOI), otherwise you may find this changing significantly, of course, once you enter Duna's SOI, make sure to set it to 12,000-12,500 again, this allows you to spend very little fuel to circularize your orbit in a stable orbit with an apopsis between 50-100 km.

One thing you can do, is avoid a "direct ascent" from Duna. I've gotten better at orbital rendevous (I don't use mech jeb, or any addons like that), there is no sense in taking the fuel you need to get from Duna's orbit to Kerbin, all the way down to Duna, and then back up again.

Leave that fuel in orbit around duna, and simply link up with it again after getting back into duna orbit.

What I did twice last night was not optimal - I had way more fuel than I needed (using 2x nuclear engines), and I took 2 command pods (the 3 person capsule, and then a 2 person lander) - extra weight, I could have eliminated the 3 person pod and put on a light probe core.

You could easily just leave a fuel tank with a docking port - no need to make it a fully functioning ship like I did (with its own pod, engines, RCS, etc), or maybe a fuel tank with a small probe core and some fixed solar panels, and docking port.

My 1st lander used 2x flt-400 tanks, and that flat rockomax tank (I think it holds the same amount as an FL-800?), and 3 aerospike engines with the two outer under the fl-400 tanks, feeding into the center (making it a 2 stage asparagus design).

The aerospikes were not chosen because the atmosphere was any concern, but because they are short, they have a much better thrust to weight than the nuclear engines (which were left in orbit), and they have a good ISP (the highest non nuclear ISP is 390, which they have and they have the higher TWR of all the 390 ISP engines)

I actually came in an aerobraked retrograde, and forgot to put separators, so I ended up taking the whole thing into orbit again as an SSTO - a retrograde orbit, because thats how my interplanetary stage was orbiting. It made it into orbit, despite carrying as deadweight 2x aerospikes (only the 3rd was going for the last part due to the fuel lines).

So the 2nd time, I used only 2 aerospikes, and instead of the central aerospike, I put a little rover underneath for the kerbonauts to use (clearance was sufficient).

It landed at a higher altitude (only used about 1 second of burn prior to touchdown), got into a prograde orbit without issue, linked with the interplanetary stage (nukes and ions), and since I had brought way more fuel than I needed, I refueled it, and left it in orbit, transfered the crew to my interplanetary stage, and waited for the return window.

Now for all future launches, I won't bring a lander, I'll just bring fuel for my SSTO lander orbiting duna, and whatever payload I want to soft land.

The 2nd time, I also launched my interplanetary stage, and lander stage seperately, and did a rendevous in kerbin orbit - no fuel savings there (its actually a bit of a waste), but the rocket launches were much more manageable than the very complex and high part count behemoth I launched the first time.

Using this method, I need to scale down my launchers (7 large orange tanks, asparagus staged), as I basically brought over fullly fueled orange tank's worth of fuel all the way to duna - ie, my central orange tank - the orbital insertion stage had plenty of fuel left for both the lander and the interplanetary nuclear/ion "tug", so I shut off their main engines, and took them to duna too, I also left them there (like I've done with many fuel tanks around the mun and minmus), I'm now in the habit of over-building my rockets, and when its time to head back to Kerbin, I guess how much fuel I need to get back (ussually with a good margin, and my main stage normally has ions, so I can ussully get it home if I run out, though it may take a while), and leave the rest in orbit with a small docking port for future use.

[Death Engineering]

As mentioned above, setting the conics patch mode to '0' should make adjusting your encounters from great distances much more precise. With practice, using only RCS, you can fine-tune your Duna encounter while still in Kerbin's SOI. Of course, it will shift slightly as you change SOI's, but you can at least get it really close for very little delta-v.

This 'Apollo-style' Duna mission has no large diameter parts (except the Cupola and Lander can and their decouplers/adapters) and still has plenty of delta-v for fudging around with encounters. Keep your craft versatile (read: docking ports and extra delta-v) and you will have no problems.

Edit: And for getting off of Duna and back into orbit, don't underestimate the "free" delta-v you get from the planet's rotation.. as long as your CM is orbiting Duna in a prograde orbit.

Edited February 17, 2014 by Death Engineering

[technion]

I just put this picture in another thread..

You can't see the lifter in this picture, but this managed to get to Duna, land and return. There's a probe which allows the transfer stage to stay in orbit and rendezvous with the lander after reorbit.

A large portion of my weight is in TAC add-ons and parts for deadly re-entry, so players without these mods should have no problems.

[saamoz1014]

Thanks everyone for the quick replies. I've decided that I'm going to go for the space station/orbtal refuel technique. This is going to be my first time docking, and I'm still working on it, but I won't bother you here anymore.

[radonek]

Well, docking can be tricky but has one major advantage: you can practice at LKO until you feel sure about it. My advice is to start small – design smallest possible ship (mk1 pod, panel, chute, small retro engine, decent amount of monoprop. and, of course, docking port) and have some fun with that. Once you get used to things, you can try with real (read: heavy and unwieldy) ships.