Mothership Designs?

[Frozen_Heart]

I'm looking into building a proper mothership in KSP and I'm wondering what is required for it to work? Is all that is needed some fuel, NERVAs, and docking ports to put landers or bases on?

Can anyone post their designs for influence?

Edited January 22, 2015 by Frozen_Heart

[Merandix]

What would you call a 'mothership'? Do you mean a large interplanetary capable craft that 'tows' several smaller ships? Not that I'm qualified to show anything, with being addicted to (space)-planes and having properly leaving Kerbin's SOI still on my to-do-list

[PrivateFlip]

yip. That's all it needs, NERVAs can be replaced by other engines of course . Don't forget solar panels.

-Things you might want to take into consideration is modularity. Instead of having all fuel tanks directly attached to the "engine block", consider adding them instead with docking ports. This way you can enlarge or swap your fuel compartment for different mission. The same goes for every other functional part of your mothership. It's a shame you if have to build a new one from scratch when you just want a larger crew compartment/need more docking ports/require a stronger engine cluster.

-Take into consideration re-usability. If you use decouplers to dump empty fuel tanks, you can not replace them on future mission. Docking ports would allow this.

-Another choice to make if you want your engines to be pushing or pulling your vessel. You can place your engines in behind your mother ship in a conventional way, especially if your using standard sized docking to attach landers and stuff, can make your vessel wobbly. Placing your engines in front makes this less of an issue, but puts constraints on the future design of the rest of the mother ship since the engines 'backblast' can not of course not be obstructed.

-Think about calamities. For example

Does the mothership has it's own command module or probe core if my lander doesn't make it back to the right orbit?

What happens if one of my landers gets wrecked on the mission? Would it unbalance the mothership in a way which would prevent it to return? Can I reconfigure the lander placement to restore balance or dump another lander and still have enough crew seats to get everyone home? Can I potentially pick up the crew of a crashed lander with another lander? (You won't always bring more than on lander, but when you do, sometimes making small changes to your design could enhance the options you have in case of a mishap.)

[Frozen_Heart]

I'm quite practiced at building one launch designs. Is it possible to have a mothership which isn't modular but can be sent up in one go? Then I can place landers, rovers, etc on docking ports on the side and front. Everyone else's motherships seem to be lots of sections.

[juanml82]

How about something like this:

Using the radial ports while firing the engines will require symmetric stuff docked to it, unless the docked stuff is light. There is also a risk of collision, but it can be solved by mounting the radial port on struts. The solar panels can be replaced by NRGs, which are probably a better idea in a mothership, as solar panels can be blocked by other pieces. The top docking port can be replaced by a 1.25 meters docking port depending on the fleet configuration.

[PrivateFlip]

I'm quite practiced at building one launch designs. Is it possible to have a mothership which isn't modular but can be sent up in one go? Then I can place landers, rovers, etc on docking ports on the side and front. Everyone else's motherships seem to be lots of sections.

What I meant with modular was the ability to change or expand it later, not whether you send it up in one go or assemble in orbit. In that sense having a mothership which is modular does not exclude it from being sent up in one go.

A good trick for sending up a large mother ship in one go would be to empty most of the fuel tanks of the mothership itself. That way the mass will be a fraction of what it would otherwise be. You would then of course have to refuel it in orbit, but it might beat the alternative of building a gigantic launch system to launch a much heavier fully fueled mothership.

[Vanamonde]

Not sure if this is what you have in mind, but they are what I use. (Downloadable.)

[Frozen_Heart]

Not sure if this is what you have in mind, but they are what I use. (Downloadable.)

Thanks. I'll have a look at them.

A good trick for sending up a large mother ship in one go would be to empty most of the fuel tanks of the mothership itself. That way the mass will be a fraction of what it would otherwise be. You would then of course have to refuel it in orbit, but it might beat the alternative of building a gigantic launch system to launch a much heavier fully fueled mothership.

I've been launching fully fuelled as my launchers can take the needed payload so far.

It seem I'd been overcomplicating things to start off with. Building a mass of struts, girders, fuel line, staging, and docking ports that wobbled around and ate up the part count.

It seem a simpler design seems to work better and I've got some small ones working now.

However Rune mentioned that pusher designs are often unstable. I've not had problems so far but is it worth switching to a puller if the payload gets larger?

[Azimech]

Funny ... I wanted to paste a link but now I see you've already found my ships :-)

[Laie]

It seem a simpler design seems to work better and I've got some small ones working now.

I've given up on large designs like Vanamondes -- missions come in all sizes and a big tug will be way too big for many, and too small for some. My standard tug these days looks more like Juan's, but with only two engines. If I need more thrust, I also need more fuel. One tank & two nervas scales well. However, managing the tugs can quickly become a major time sink. They need to be returned (costs fuel), brought into a stable orbit, RV & docking (costs time)... also see this discussion.

My latest Jool-5 required four tugs. One the one hand, I'm proud of my infrastructure and like to use it; on the other, the logistics took up a whole evening.

However Rune mentioned that pusher designs are often unstable. I've not had problems so far but is it worth switching to a puller if the payload gets larger?

It's not a question of size as such, but of wobbliness; if your vessel can bend and flex a lot, pushing will make that even worse. For practical purposes, the more docking connections you have, the worse it becomes. If the payload comes up from Kerbin as one strutted bundle, and you only detach pieces as needed, it can be any size without problems. Having the reaction wheels distributed evenly across the vessel's mass also helps. In my opinion, if wobbliness becomes so bad that push vs. pull can even be a topic, that is indicative of a planning/design error.

[Azimech]

There is another solution: having mass distributed both in front and behind the engines. And wobbliness needs to be no issue if you build a tug in a shape that gives it a strong shell.

If you want to see what I mean, take look at the link in my signature.

Edited January 22, 2015 by Azimech

[Frozen_Heart]

There is another solution: having mass distributed both in front and behind the engines. And wobbliness needs to be no issue if you build a tug in a shape that gives it a strong shell.

If you want to see what I mean, take look at the link in my signature.

Might have a look at some of those designs. They're sleek, reusable and single stage which if the same as what I'm doing. The internal cargo bay is cool as well though my current designs are much too small to do that.

[wibou]

Hmm a modular and super-heavy mothership that can dock several smaller rockets/landers sound like a neat idea. Make sure to post your results back here, it sure would be interesting

[Frozen_Heart]

Hmm a modular and super-heavy mothership that can dock several smaller rockets/landers sound like a neat idea. Make sure to post your results back here, it sure would be interesting

I'm starting small to test my building skills. I'll gradually increase the size as I get more practiced. I've got a few in the WIP thread but honestly they're a bit small and short ranged to be called motherships. :/

Hopefully they'll improve in time. Still a very long way off having something capable of the Jool V challenge.

[cicatrix]

You don't really call it a mothership since nearly all of them are nothing more but tugs. A mothership is something else - a ship that is nearly autonomous - this means ISRU of course which is not yet possible with stock. A true mothership can be autonomous for indefinite periods of time, capable of a grand tour. A tug is quite another matter - its sole purpose is to take you to the destination system and back.

[sdj64]

You don't really call it a mothership since nearly all of them are nothing more but tugs. A mothership is something else - a ship that is nearly autonomous - this means ISRU of course which is not yet possible with stock. A true mothership can be autonomous for indefinite periods of time, capable of a grand tour. A tug is quite another matter - its sole purpose is to take you to the destination system and back.

Not true, a mothership could require refueling. But, I would say a mothership has to have crew, which a tug does not. The crew can go down to the surface of a planet on landers and some might stay on a permanent base but the majority stay with the ship. I'm also partial to internal hangars and reusable landers or shuttles as a requirement for my motherships. You can fake autonomy by sending fuel tanks ahead of time and do a longer multi-destination mission with refueling, or just give it enough dV that it doesn't need to refuel (my recent Jool-5 mothership had more than 8000 dV, and I have one in my modded career with Near Future that has more than 30000. It could do a grand tour without refueling or ISRU, except for not having landers for Eve or Laythe).

Frozen, even with your one that's on the Spacecraft exchange (nice looking ship btw) if you had a Mk2 hangar you could put a Mun or even Duna lander inside. Or you could have auxilliary satellites/ unmanned landers for the "science from space" and temperature/gravity/seismic contracts. For example, you're going to Duna and you accept some of those contracts on Ike for your unmanned lander to do.

[Frozen_Heart]

Finding the problem with a single stage mothership now is that for a certain TWR ratio there is a max amount of dV you can have. And that dV isn't very high for a mothership.

It ended up as something like 13km/s for a TWR of just 0.2. That's with just fuel and a single engine so everything else reduces it much further. I'm struggling to get above 7500.

[Laie]

If you take one LV-N and try to stack an infinite number of fuel tanks on top of it, your delta-V will asymptotically approach something on the order of 17km/s. If you want to get more, you'll have to use drop tanks that will be discarded when empty, reducing your ship's dead weight. This is the rocket equation at work, there's no way around it.

However, 7km goes a long way, and for interplanetary missions I find 0.2 to be still acceptable. Your longest and worst burn will probably the one that takes you out of Kerbin, but even for that purpose there's people who would claim that 0.2 is already way overpowered. As you seem to feel different, I suggest you consider a "kicker" stage that will improve your TWR for the duration of that first burn, and is then discarded.

About delta-V, what do you want to do with your mothership? Jool and back is ~3300m/s; if you also want to bring it into an orbit around every moon, you'll need about 4500m/s-5000m/s. So a Jool-5 mission can technically still be done without dropping tanks. If you want to make a grand tour, however, you're out of luck.

[Frozen_Heart]

As you seem to feel different, I suggest you consider a "kicker" stage that will improve your TWR for the duration of that first burn, and is then discarded.

I usually aim for a TWR of 0.5 - 0.7 and I can get 5-6km/s out of that. For slightly larger ships 0.4 - 0.3 is ok and I can get 7-9km/s from that. I'm not a patient woman. :/ That kicker stage is a good idea but damages the reusability of the ship.

Jool and back is ~3300m/s

For some reason I thought Laythe return was about 10,000m/s. Good to know its a lot less!

[EdFred]

Probe + large SAS + HH Pod + Fuel Tank (orange or half orange) + 2-4 radial mounted LV-Ns on giirders + Clampo-Sr's at each end along with a 4k battery and some solar panels is what I go with. I try and only use 2.5m parts anymore.

[Rakaydos]

For some reason I thought Laythe return was about 10,000m/s. Good to know its a lot less!

That sounds more like Laythe Direct- Kerbin launch to Laythe surface and back.

[Laie]

That kicker stage is a good idea but damages the reusability of the ship.

First, let me repeat that reusability is nice, but actual reuse means a lot of additional maneuvering. Be sure that you like this before you commit a lot of infrastructure.

That said, nothing prevents you from reusing the kicker stage as well. See my own Jool-5 entry for an example (the design has flaws and I'm doing it wrong, but it's an example). Returning it immediately (like I did, wrongly) will cost more dV than the initial burn. It's not that much fuel, on the face of it, but toy around a little with locked tanks and you'll see how it impairs your outgoing TWR.

The better way is to still let it go off on a long voyage, but to reduce solar apoapsis just a little, so it's trip around the sun will last exactly N full years, give and take only a few days. That way, when it comes back from it's round trip and touches Kerbin's orbit again, Kerbin will be there for capture. It may be away for years, but it will still be home before the mothership comes back. If you budget 600m/s for the whole recovery thing, that will be plenty.

[AbacusWizard]

In general you want plenty of fuel, efficient engines (probably atomic, though ion can be fun if you're willing to turn on the engine and go do something else for half an hour), crew quarters, a mobile lab if you're going to do any science, batteries, solar panels of your choice, maneuvering jets, monoprop tanks, and enough docking ports for whatever shuttlecrafts or probes you're bringing along. Beyond that, it depends on your intended mission.

One very important thing to keep in mind is that your fuel tanks need to supply not only the mothership's ∆v needs, but also refueling for any reusable landers. Even an ion-powered mothership needs some rocket fuel tanks if that's what the landers run on.

I generally build interplanetary motherships for ongoing missions of exploring strange new worlds, seeking out new life and new civilizations, boldly going etc etc etc rather than a specific there-and-back-again jaunt, so I don't really start with a specific ∆v budget in mind; instead I pack as much fuel on as I can reasonably get up to orbit and then send tankers to rendezvous with the mothership at its current location whenever it runs low on fuel. If I'm in a planning-ahead sort of mood I send a tanker to the mothership's intended destination ahead of time so it's already there when the mothership needs it.

Example for your perusal: This is the Phoenix III, a nuclear-powered mothership from two versions ago that I took to Duna, Ike, Dres, all the moons of Jool, and back home. It has eight orange jumbo tanks, each with a nuclear engine on the aft end and a medium docking port on the fore. The central column includes two giant monoprop tanks (primarily for restocking the landers, which need it to re-dock effectively), a probe core for autonomous operations, a large docking clamp (to dock with tankers and refuel), a mobile lab with all the latest science equipment, and a cupola for the pilot. The crew section (lab and cupola) is detachable and has parachutes and landing gear in case of unplanned lithobraking. During its original mission it carried one single-occupant crew lander and three uncrewed probe landers; my general method was to enter orbit around a planet or moon, send a probe lander down to the surface and bring it back up to make sure it could be done with the available ∆v, and if so, send down the pilot in the crew lander. Ended up leaving probe landers behind on Laythe and Tylo; everything else was able to make it back up.

Unexpected bonus: I realized that by removing the crew section (leaving just the tanks, engines, docking clamps, and probe core), I was left with a reasonable utility tug for interplanetary hauling. Over time I launched five or six of these P3-Class Atomic Tugs and used them to carry orbital stations and ground outposts to Gilly, Duna, and Laythe. Good times.

[Azimech]

Working on a new concept for discarding empty tanks. Yes, a lot of parts. But the ship lost 3.2t and 24 parts with the touch of a button.

[Frozen_Heart]

Returning it immediately (like I did, wrongly) will cost more dV than the initial burn.

Maybe using it to get right to the edge of Kerbin's SOI, and aerobraking it when it comes back around so its home again with very little extra dV? According to the wiki that could save about 940m/s if done right. Not much but it improves the TWR and is still a welcome boost.

Working on a new concept for discarding empty tanks. Yes, a lot of parts. But the ship lost 3.2t and 24 parts with the touch of a button.

That looks pretty cool. Are you using sepertrons to get them away from the ship?