craft planning

[Starwhip]

Basically:

1. Pick a destination

2. Build a lander

3. Test lander

4. Build a CSM/Unmanned tug

5. Test lander/CSM combo

6. Revive Jeb...

7. Dig up some kind of lifter craft or rebuild one.

8. Stick it on the bottom of the payload.

9. Forget the launch clamps the second I hit "launch"

10. Go back to VAB and add clamps, forget sepatrons, add sepatrons.

11. Lift off about 10 meters off the ground

12. Explode

13. Screw the hour and a half of my life I just wasted and walk away.

This is not really how I do it, steps 11-13 are just for fun, usually it gets where it's supposed to by step 10.

[7499275]

I usually just have an idea in mind of what needs to be done and then build on the spot with a rough idea in my head.

[Captain Sierra]

My design methods for a space station:

1) craft concept in my head

2) refine every detail to the point where it can be built exactly to spec in KSP, still in my head (I'm freaky good like this)

3)build entire station in VAB

4)use subassemblies and SelectRoot to break into subassembly modules

5) start anew, grab some random part, use SelectRoot on module, delete random part

6) build an RCS docking tug for payload if it is not the core stage

7) build upper stage, test structural stability if necessary

8) build lower stage, typically a two stage Delta V or Onion staged style lifter(enough to reach target orbit and not leave any debris)

9) launch, instantly revert to VAB knowing I forgot SOMETHING (usually control for the upper stage)

10) launch, identify faults, revert, correct, repeat as necessary

11) succeed in getting orbit and docking

12) profit

[SDEngineer]

I keep telling myself I'll save a bunch of subassemblies of lifters so I can just grab one that's the right size for the job, but they keep getting out of date because I keep unlocking better rockets and fuel tanks.

Definitely the way to go. I have somewhere around 10 subassemblies ranging in capability from 15t up to 120t (all rated to 100km orbit). Saves so much time setting up the asparagus staging, fuel lines, and struts on what is generally the largest and most fragile part of the final rocket (and most prone to catastrophic failure if you forget a crucial set of struts or make a mistake setting up the staging), letting me concentrate on the payload that is actually new construction. I can't count the number of launch failures I had due to hastily constructed launch stages before subassemblies.

[goldenpeach]

I usually start with the payload and build the rocket around it.

If my payload consist of different module(e.g a CSM with a LM attached to it), I build and test each of them separately, usually starting form the "bottom"(in the previous exemple, I would start with the LM and build the CSM it need to do the mission).

[GamerMitch]

I just pretty much let my hand, my mouse, and my mind run on autopilot, and almost everything I make works

[KilrBe3]

Still pretty new to KSP , but not for the first time. But this time around with mods and whatnot, so its a lot better!

1) Brainstorm what I need, (Kethane Probe? ISS Part? Lander for Mun? etc etc)

2) Go back and forth an hr on just the payload part I need/want.

3) Throw it into a subassembly.

4) Load LazTek's SpaceX Mod

5) Put PayLoad ontop

6) Launch!

Yes yes, my friends say that is cheap and not the intent of KSP. Last I checked, this was KSP, a Space game. My close friend refuses to put it on his KMP server, because its pretty much cheating to him. Rocket, Staging, Payload transfer, its all there. Just put a payload on top, adjust Fairings, and launch. I fully enjoy it, due to I love SpaceX Idea/Mission/Design of their rockets and command pods (Dragon). So I use it for everything, and I am still not bored yet. Why? Because I enjoy using a rocket I know is solid and works. So do my friends, expect they built them, but they still use that same tried and true rocket over and over to launch their payloads. So whats so special and cheating about mine? Nothing. Whiners gonna whine!

I mostly want to create cool things or see different planets, not sit there for 2 hrs testing a rocket. That's 2 hrs I could have had improving my ISS, going to Mun, making a Mun ISS, Scanning for Kethane, etc etc. If I do get bored, I will just make a rocket. Its not very hard, its just time consuming. If you know what you are doing in KSP, and can do everything without MechJeb, people make rockets with their eyes closed pretty much in minutes compared to me

[LethalDose]

Like a lot of the previous responses, I build top-to-bottom, but before I start putting parts together I start with a mission profile (e.g. Manned vessel to orbit & return from Duna).

From the profile, I build the vessel that will go from LKO to the target. After that's together, I build assemble the launch stages. If the mission requires multiple launches, I'll usually send the "core" up first, then, the support missions. For me, this minimizes my number of launches because unexpected issues may arise during the core launch, and the following support launches can perform their original mission and help fix what went wrong.

[Ithicatic]

At the core, my design methodology is top-to-bottom. I don't design on a mission-by-mission basis though. Instead, I try to design for a set of missions; e.g. Mun Missions, low-grav no-atmosphere missions, atmosphere land/re-orbit missions, etc. I always set up a fuel station in any SOI I intend to land/re-orbit in multiple times, so often I'll design payloads to fit the specific environment of that SOI. Like my Duna VTOLs won't be going to the Jool system nor Eve, since they can't return to orbit from the surface of every body in the SOI, but they will probably be used in any Dres missions. I also have a LKO launch vehicle series that covers a range of payloads from 5 tons to 100 tons (and a couple untested launch vehicles that theoretically can lift up to 150 tons), so I don't have to redesign the lifter for every launch.

HOWEVER

In Career Mode, I design a new series of launch vehicles and payloads from the ground up every time I complete a tier or unlock major advancements (like LV-Ns). I do this because I know that I'll be redesigning my work-horse lander/CSM anyway. I'll make the launchers first (bottom-up approach) before any payloads, work out the bugs, and test the actual carrying capacity instead of relying on just my pre-design number-crunching. Once I have my launch vehicles, I know what kind of tonnage range I can work with, and re-design my payloads within that limit.

[Rainbow-Dashie]

I build what I want to land/put in orbit, the space station core, rover, munar lander, etc. Then i pretty much put enough rockets underneath it until i think it can reach it's destination. If it doesn't I will add more rockets, if it still doesn't work, I make the payload a lot smaller and lighter. if that fails, I give up, and move to the next stupid idea until i decide to come back to this!

[SkyHook]

Well I ask myself a series of questions:

1. What do I want to do?

2. Do I have a craft that can do this?

3. SPH or VAB?

4. Reusable?

5. Is this even sane?

These questions allow me to start the process of completing the mission. I have builder's block right now so there isn't much being done lol.

[FITorion]

trial and error.

some ideas from watching others. I would not have thought of asparagus staging myself.

[ArmchairGravy]

1. Mission

2. Payload(s)

3. Boosters

Kerbal Engineer really, really helps with 3!

[Levelord]

Clever usage of the VAB and subassemblies...

[Starwhip]

/\ THAT is really cool!! Love the pan/rotate shots, looks like it wasn't made in KSP!

I wanna make a thing like that... although I can't make a VTOL, and SSTO's are elusive, but not impossible. I think I just make them too big.

So here is my REAL craft design process.

Think of a place I'd like to go and what I want the craft to do.

Make several prototypes of all components.

Make more prototypes of all components combined.

Make a final design.

Launch and achieve SCIENCE!!!

[mellojoe]

Throwing something together then spending ages refining it.

Mostly this.

Often times I have a basic idea of what I want to do. Something like: "A triangular shape with a useable docking port on both the bottom and the top." And then I have to decide if I want this particular craft to be manned or un-manned. And how many (if any) passengers. But then its back to just slapping stuff together and seeing what comes about. I litterally spent 3 hours last night trying to design a shuttle to from a Mun base to an orbital base. And I probably threw out the first 3 designs, settling on something much different than I originally intended.

[magnemoe]

I tend to build the lander or payload first, later I have tried to also use the lander for Kerbin landing. As most missions in carrier has been two man missions with an one man lander I have used the Mk1 pod as an escape and return system for the second kerbal while one land with the lander.

For complex missions I test each lander before importing adding to the framework, then add fuel as needed.

This is an Jool grand tour mission in 0.22, land on all moons and drop two probes into Jool. An 0.23 version will be more complex as I will have to replace science and goo containers.

To keep part count down I will use extraplanetary launchpads and build the landers at location, it don't save any weight, it might save some room even if rocket parts are bulky but saves a lot in part count.

[KvickFlygarn87]

(somehow, I always get great ideas just before I fall asleep).

Glad I'm not the only one.

[Commissar]

generally, i'll build the payload, and if it's a lander, I usually test the landing legs and such, make sure that is in order, then I build the rocket itself, the process of that really depends on the payload and destination. then I try the mission, and frequently encounter a problem, such as bounce and/or descent targeting.

stuff that will be permanently in space can't really be tested till it's in orbit anyways, so I fly it till i have a problem. if I do, i just deorbit the craft (after removing any kerbals) ... and refine the craft.

[Randox]

My launchers are pretty standard. I tend to build to a certain range, and everything leaving Kerbin is usually destined either for a medium-high Kerbin orbit, or strait to minimus where my staging area lives (Kethane mining).

For ship design, I mostly use my orbital tugs as a fleet backbone. It's a very capable yet slightly minimalist design. On their own, they move payloads from planet to moon and back (or moon to moon), usually with the understanding that they require refueling at each end of a trip. The idea is that if a mission discovers an unanticipated need, rather than send out a new ship to take over, I can often just design a new module the existing ship can use and send that out.

[Captain Sierra]

I just started a grand tour mission (link in sig, you know you want to click it . . .) so here's my design strategy for that

1) decide to do a grand tour

2) open VAB

3) Build entire tour ship, starting with the drive stage

4) scrap the entire utility section, rebuild it because I didn't like it

5) decide what needs to come with me

5) redesign utility section as necessary

6) design landers, SSTOs, whatever crap is coming with me

7) go over the whole thing onceover with a fine tooth comb

8) begin launch and assembly of the parts I'm satisfied with and am certain they are what I need

9) continue tweaking rest of it

10) launch when satisfied

11) add fuel and crew, time warp to Jool window, execute 10 minute burn with 12 NERVAs pulling 200+ tons

That about sums up my megacraft design process. Sometimes I end up hung up on 4 and get sandbagged there for a while.

[Landge]

I'm currently designing a new Mobile Base. Which comes equipped with:

6 Coupola's

2 Hitchhiker Modules

1 Mobile Science Lab

plus all the sciency related parts and apparatti...

Also retooling my Fuel Truck and various craft for on the ground refueling.

For 0.23 I'm keeping my Fuel Truck's side mounted rockets when I need VTOL. (She can't handle steep grades.)

Here is my Kethane Rig for refueling the Fuel Truck.

Current mods in use:

Chatterer

Kethane Pack

Edited January 29, 2014 by Landge

[brdavis]

Interesting. I'm still brand-spankin'-new to all this, and every time a new version comes out, I start a brand new career mode (hey, we're supposed to be testing, right?!?). So i've not gotten too far, but just completed my first Kerbaled mission to Duna.

1. Set goal (get to Duna and back with a live Kerbal; sanity optional)

2. Sit down and work out exactly how that's going to work, backwards:

--Kerbal atmospheric entry?

--Duna-Kerbal transfer stage?

--Duna ascent stage?

--Duna lander?

--Kerbal-Duna transfer stage?

--Kerbal heavy lift vehicle?

I work out all the delta-v's, required thrusts, fuel loads, docking & reconfiguration procedures, all at this stage.

3. Go to the VAB and build first part, testing it unmanned

4. Refine until it works, then back to step 3 for the next part. Repeat until finished.

5. Launch it and perform the mission, discovering something like "oh yeah, I really did need batteries" or similar.

6. Success. Or, spectacular, very well-researched failure. Whatever.

For me, planning it out is as important a part of the game as anything else - because for me, that's fun. when I did the mission and got the Duna ascent stage to dock with the interplanetary transfer stage that had been left in orbit, that was wonderful… but the fact that I only had just enough reaction mass to do it is what made my day .

[Landge]

That goes for me too. With the exception of calculating delta-v requirements. I haven't used MechJeb or Kerbal Engineer. I'm just now getting around to adding Kethane Pack.

Be sure to post pics of your builds!

Here's more on my latest build...

After field testing the rover I copy it over to the VAB to add boosters... "More boosters you say?"... Yes! MOAR BOOSTERS! Started out with 6. Then 8. Finally settled on 10 plus a few SRBs (6) for good measure.

When all is said and done the ship totals 800 parts and weighs in at 1414.5 tonnes. Not the largest I've ever built but still...

Getting into LKO...

Making the first gravity turn at 10,000 m...

Second gravity turn at 30,000 m...

In stable orbit over Kerbin...

It has enough fuel left in the transfer stage to make it to Mun or Minmus. But with the Clamp-O-Tron Sr. mounted on the back I have the option to add extra fuel tanks.

Now where should I take this thing? Anyplace airless will do because I forgot to add the blasted parachutes!!!