Opinions about craft size

[MisterKerman]

I've always admired smaller minimalistic designs as long as they are fully equipped with all the comforts that make space exploration easier for the operator. (Not the cringy low part count/lowest tech level designs.)

But there comes point where you need to bulk up your design to physically move equipment/craft you plan to bring with you, as well as the heavy fuel required for the expedition. (Think Jool motherships.)

I feel like ideally as long as the fuel margins are forgiving enough, smaller ships will always be ideal. I struggle greatly not over-engineering craft. It just feels wasteful to be lugging so much extra weight around, however I've never left Kerbin's SOI outside of training missions for astronaut XP to level them up. I don't have a real idea what size is the most useful; only which size is most useful for visiting Kerbin's moons.

I was hoping to get some opinions from members about when/why they build big and how hard they struggle themselves to keep efficiency a priority when designing interplanetary craft.

[MechBFP]

14 minutes ago, MisterKerman said:

I've always admired smaller minimalistic designs as long as they are fully equipped with all the comforts that make space exploration easier for the operator. (Not the cringy low part count/lowest tech level designs.)

But there comes point where you need to bulk up your design to physically move equipment/craft you plan to bring with you, as well as the heavy fuel required for the expedition. (Think Jool motherships.)

I feel like ideally as long as the fuel margins are forgiving enough, smaller ships will always be ideal. I struggle greatly not over-engineering craft. It just feels wasteful to be lugging so much extra weight around, however I've never left Kerbin's SOI outside of training missions for astronaut XP to level them up. I don't have a real idea what size is the most useful; only which size is most useful for visiting Kerbin's moons.

I was hoping to get some opinions from members about when/why they build big and how hard they struggle themselves to keep efficiency a priority when designing interplanetary craft.

The more efficient you build a craft, the more DeltaV it can have, and the smaller it will be, which makes it easier to launch without exploding.

However that is only necessary for very high DV missions (I.E an Eve land and return craft). Otherwise building super efficiently is a waste of time unless you are super strapped for cash. 

[Pthigrivi]

I would say the biggest driver of bigger ships for me is immersion. Of course, I could stick one Kerbal in a MK1 lander can (or command chair) and send them off on an 8 year Jool mission by themselves, but I tend to model things with at least some kind of sense that this is a real mission with enough space for a reasonable crew, often 6-8 kerbals for interplanetary missions. Nertea's SSPXR has some incredible hab modules for this kind of thing. Thats not to say the drive systems and mission profile can't be designed really efficiently around that, but I like an underlying payload that I could imagine the little dudes living in for a few years. 

[MisterKerman]

@MechBFP

There's going to be times where you need high TWR in order to complete maneuvers within a short enough time. I know it scales up and down subjective to payload requirements, but I tend to find going bigger more forgiving. Just trying to get some replies to see if I'm correct in this way of thinking for most cases!

I'm really clueless when it comes to interplanetary. I've never even thrown something out to see where it goes like garbage out a car window.

[steuben]

There is the min-max approach. And it does have it's adherents. And I use it, every once in a while.

But, when I'm playing career I tend to go for the dv+ approach with adjustments for style. 'cause Doc Brown was right.

In sandbox ... style is jumping up and down on every other consideration... with cleats on.

 

[Loskene]

Well in real life the square-cube law means building bigger is practically always better in terms of efficiency, capability and redundancy, typically in the context of crewed spaceflight. If you can afford it anyway.

The conditions inherent to KSP as a limited simulation with its own particular quirks muddy the waters here and make small a more viable option, so it's really just a facet of how KSP differs from reality.

[MisterKerman]

I just want to see what people tend to do. I'll probably start small and fling a satellite at Duna attempting a capture. Not going to mess with Ion engines yet though. Seems less straightforward.

If they give a great reason why they do it in the process I'd find that very helpful. Seems like aside from payload requirements it seems like the consensus is bigger is better.

[Pthigrivi]

30 minutes ago, MisterKerman said:

I just want to see what people tend to do. I'll probably start small and fling a satellite at Duna attempting a capture. Not going to mess with Ion engines yet though. Seems less straightforward.

If they give a great reason why they do it in the process I'd find that very helpful. Seems like aside from payload requirements it seems like the consensus is bigger is better.

Not necessarily. The payload is your choice depending on what you'd like to do out there, but when it comes to your delivery systems you should really be keeping things as efficient as they can be leaving a little extra for the occasional unexpected inclination burn or inefficient landing. If you aren't already you can use the dV map (the easiest way) or a launch window planner (the more accurate way) to estimate how much total delta V you'll need. Keep notes on how much you'll need for each leg and then design your stages to meet that (with a little extra). For your first few interplanetary missions you'll want to make sure you're leaving at efficient transfer windows and probably dropping your transfer stage before you attempt to land your probe so you're not hulking more weight than you have to down to the surface. Waste tends to grow exponentially over high dV missions, so packing too much fuel for a Moho lander could double or triple the mass of the mission off the pad, which adds cost in career but hassle even in sandbox. 

Edited March 29, 2019 by Pthigrivi

[Superfluous J]

I'm kinda in both camps. Sometimes I want to be super efficient. Sometimes I just want to get that thing into space so I toss a couple huge fuel tanks and some mammoths under it.

[Foxster]

If you want to be efficient and have smaller craft then it's usually all about payload. 

You often at most need just to get a man in a can there and back. But for some reason folks want to add a couple more crewmen, a rover, parachutes in case Eeloo suddenly gets an atmosphere, and a ton or two of other junk. Every bit of extra payload balloons the whole craft greatly. 

KISS. 

Edited March 29, 2019 by Foxster

[Aegolius13]

1 hour ago, Loskene said:

Well in real life the square-cube law means building bigger is practically always better in terms of efficiency, capability and redundancy, typically in the context of crewed spaceflight.

On the flip-side, the square-cube law generally makes it harder to aerobrake/reenter, so that one can be a double-edged sword.  

---

Hopefully this is not a dodge of the question, but I view size as more of an output than something to aim for, one way or the other. I.e., I try to design ships from the top down, trying to be efficient with mass while leaving reasonable margins.  I guess that generally results in craft at the smaller end of the spectrum, but I have no problem with launching something big if it's a sensible way to complete a mission.  

The exception for  me, though, is mining rigs and fuel tankers.  I have a strange obsession with making absurd monstrosities like this.

 

[Superfluous J]

30 minutes ago, Aegolius13 said:

The exception for  me, though, is mining rigs and fuel tankers.

I know how that goes. "But wait, what if I need to totally fill TWO Jool 5's at the same time?"

[Loskene]

Mining rigs must always take up an excessive amount of surface area or they don't look "miney" enough.

[MisterKerman]

Ultimately I plan to build a 2.5m mothership of sorts. I feel like that's a good size depending where my destination is. I'm not sure how subjective craft size is or when it's deemed necessary to build large. Every Jool ship I've seen has been enormous.

[Loskene]

3 hours ago, MisterKerman said:

Ultimately I plan to build a 2.5m mothership of sorts. I feel like that's a good size depending where my destination is. I'm not sure how subjective craft size is or when it's deemed necessary to build large. Every Jool ship I've seen has been enormous.

I think it's really up to your preferred playstyle whether you want to go micro or macro, there are equally viable options and opposite but equal challenges involved in both.

I like to build large, mainly for the air of realism without going full realism overhaul, and so one essential mod I use in career mode is USI life support, specifically for its habitation mechanic that forces you to build appropriately spacious craft for long duration missions. It all scales very nicely by crew count and expected return time.

I'll put an example craft from my old 1.5 career in a spoiler below, of a simple mission to orbit Duna and return home safely, meaning sending life support and most importantly comfortable habitable volume to sustain a crew of just 3 for more than a year or two. Because of the mass penalties imposed by the life support and sanity concessions, the vessel size balloons from what would normally be a pod-on-a-terrier kind of mission to a full blown interplanetary mothership with tight delta-v margins. The ship had to be launched in 3 sections, the main command, science and power module which could also serve as a station (since it's bloody expensive and I wanted my money's worth), the propulsion module to get it there and back again, and finally a Soyuz to bring the 3 kerbals aboard and also serve as a lifeboat and their return capsule for the duration of this soviet-themed Duna exploration. I could've left the Soyuz in Kerbin orbit to save mass, but when docked it provided free habitation time with the extra volume and even after refuelling the ship with a tanker launched to Duna at the same time, I still couldn't fully guarantee capturing back into Kerbin orbit afterwards. In fact it was so tight on the return leg I had the crew bail into the Soyuz and undock upon reaching Kerbin SoI, so that the mothership would be just barely light enough to capture into a loose orbit. Worst case scenario I'd've had to recover it from Kerbol orbit at some point, but having the Soyuz saved me a salvage mission, or nearly caused one, who knows. You can see by the ratio of fuel tanks to equipment it's not got the best mass fraction, I was limited by the launchpad tolerance and didn't want to do more launches to fuel it, it already took 2 after orbital assembly, plus the cost of the tanker sent to Duna meant the mission was really making a dent in the budget. Going all the way to Jool without getting top tier parts and KSC upgrades would be quite a challenge under these circumstances I think. Definitely worth looking into for anyone who's never tried life support before, make sure whichever one you pick has habitation though, food/water/oxygen alone doesn't give the same effect.

Spoiler

Hab says indefinite there because it's the vet crew but I did intend to reuse this so it's built to spec. Don't mind the expired EC on my other station, the game doesn't calculate it when the vessel's not loaded.

Meeting the tanker in high Duna orbit after burning some extra fuel doing a close pass of Ike for the science. We had 5 days to kill waiting on the fuel truck and plenty of snacks to spare.

Bye old friend, hope we see you again some day (you won't, I started a new save)

"Have we landed yet?" "I don't know, get out and have a look around"

Edit: Oh wait, those last 2 are from when I brought the LKO station crew home before they got back. They were the next screenshots in the folder. Oh well.

Edit 2: I just found I have old launch and orbital timelapse footage from this mission and one of the earlier stations that I never did anything with, and it's pretty decent. Might as well slap together an edit and see what we get.

Edited March 30, 2019 by Loskene
stuff

[M_Rat13]

Honestly if are worried about interplanetary missions in any way, go the rover/drone route. I'm guessing you play in science mode, as you don't need to level up in a creative save, and career mode will force you outside of Kerbin SOI to make money.

With that in mind, rovers can still earn you science, if built right. Also, if you have comsat on, you'll need to bear that in mind, and build a network around other planets, or even Kerbin if you have the relays on the ground disabled.

Just use experiments that are repeatable (don't require an engineer to reset), and a powerful enough antenna to either send information through the relay network (if you need to), or straight back to Kerbin.

The main advantage is that once you land a rover, if done right, you don't need to bring it back, unlike Kerbals which are valuable. It's an interesting way to explore planets for the first time.

Drones are also useful for finding the best ore locations, if you want to mine, or deep space exploration, if you plan for a craft to stay in deep space for long periods of time (fly by all the celestial bodies).

[Aeroboi]

I always prioritize part count.

Why?

Because sometimes vessels meet in orbit, so I want to dock a 200 part craft to another 250 part craft and that is very laggy. So I try to build with as little parts as possible. When prioritizing for part count you are prioritizing for weight reduction in a way because you are also often scrapping off those extra Kg's doing so.

Mostly you only need the minimalistic. See @Stratzenblitz75 Jool 5 mission to see what minimalistic is. Does he use few parts, is his lifter lightweight, yes!
It is a minimalistic fueled design and only the needed requirements are on board.

Sure you can build a SSTO to launch 100 Kerbals on a Jool 5. But why? In career mode you usually only need 4 Kerbalnauts for career missions. So a Mk2 space plane with mk2 nose cockpit and inline cockpit could serve the career contracts for science gathering. You only need a small on board orbiter, maybe a setup of 2 rocket modules that can dock and function as a Tylo lander, science equipment and a ISRU if you want to SSTA.

Why would you need to bring more stuff? If contracts specify it, like creating a liquid fuel space station around Laythe or Jool or something like that. But then you are prioritizing towards that contract, and it's the contract that decides how big of a rocket is required to pull of the job. If you want to go bigger you are intending to launch big sized cargo for unecessary large orbital stations. There's no reason to do so other then it being cool.

I have a save with a large 10kiloton space station in Kerbin Orbit. I have a Minmus mining station with a large fuel truck that aerobrakes at Kerbin to refuel the spacestation. Admittedly I never made use of it because it is faster game play wise to launch another rocket off of Kerbin.

[MisterKerman]

@Aeroboi

I want to bring stuff with me just for fun. Relay satellites, rover, lander, maybe even a lightweight ion glider if I can find a good design to copy.

Not for science or cash due to already having my tech-tree completed and about 6 Mil of fun money in the bank. I just want to have a full fleshed adventure.

And yes, the motivation behind why I build a lot of things the way I do is for looks as well as novelty; however impractical. I'm slowly getting better at it and starting not to hate the way my crafts look/perform.

[ARS]

Having a rather underpowered computer to run KSP, partcount is really important to me, I have a maximum limit of how many mods installed, how many parts in a single ship, etc. to at least guarantee an operational mission. I love simple, minimalistic design, and there are times where I had to deal with mission that requires me to build (normally) enormous craft with hundreds of parts that could potentially become a CPU-melting contraption. To deal with this, I'm basically only strap an absolutely bare minimum modules necessary to complete the mission, and if possible, drone only. My favorite build is a self sustaining propulsion. Basically creating a closed resource system or some way to generate propellant to ensure a self sustaining vessel with the least amount of parts needed for structural purpose. Essentially a small, simplistic craft that packs potentially an infinite Dv compared to actual size of the craft needed for the mission. Normally, there's a mining, but some mods can make it easier, such as GN drive (absolutely OP infinite Dv at no cost), LBSI fuel generators (turns fuel into fuel (weird? Yes, because output is bigger than input, aka inifite fuel)), Buzzard collector (gradually generates fuel at the small cost of constant electricity), and basically nearly everything from Solaris Hypernautics (virtual particle reactors for Kannae drive, Dust collectors/condensator combo, electric-only engines...)

[Grogs]

On 3/29/2019 at 1:34 PM, Pthigrivi said:

I would say the biggest driver of bigger ships for me is immersion. Of course, I could stick one Kerbal in a MK1 lander can (or command chair) and send them off on an 8 year Jool mission by themselves, but I tend to model things with at least some kind of sense that this is a real mission with enough space for a reasonable crew, often 6-8 kerbals for interplanetary missions. Nertea's SSPXR has some incredible hab modules for this kind of thing. Thats not to say the drive systems and mission profile can't be designed really efficiently around that, but I like an underlying payload that I could imagine the little dudes living in for a few years. 

I feel largely the same. I don't run any life support mods, but I do include a good bit of extra space for role-playing purposes. For trips to Mun or Minimus, I'm OK with 1 seat / kerbal, but for longer missions I would want at least 2, or maybe 3 seats each. For my Jool-5, I took a crew of 9 even though the level 3 mission only requires 5. I put a Mk3 cockpit and two 16-kerbal Mk3 passenger modules on the mother ship, which gave 4 seats per crew member. I figure that's enough space for the ship to have sleep cubicles, a small kitchen, acceleration chairs, etc. There were also 4 other Mk2 or Mk1-3 pods between the landers, so I figured that gave them space to go get away for a while if needed. I only use command chairs for surface landers on low-G worlds.