What’s cheaper: refueling in orbit? Or launching everything at once?

[Superfluous J]

*Insert lazy voice here*

Just launch them all at once. It takes less time and effort

Ew. Those words make me sick.

Don't think of it that way. think of it this way:

Just launch them all at once. It gives you more time to spend more effort doing the actual mission part of your mission.

[Frozen_Heart]

Launching all in one go is cheaper and more efficient. However if you have a Kethane station on somewhere such as Minmus this completely changes.

[vixr]

For me, I find having a station in orbit allows me to refuel as well as stage my various craft, in orbit. I can choose whichever delivery craft suits the mission and dock it with a suitable lander. Then the craft are returned to the station to wait for the next mission. I can leave off parachutes and other things for purebred space vehicles that will never return to Kerbin. Lifting fuel and RCS is easy enough to resupply the station.

[Ruedii]

Theoretically a single launch SHOULD be more efficient, but the larger the payload the larger the rockets needed, and larger rockets might not be as fuel efficient, and the larger the rocket the harder it is to make it an SSTO.

[Toastedbuns]

Funny, that's how I define "fixing my mistakes" a lot of times

Consider the fuel station like a giant orbital eraser of sorts then

[Brainlord Mesomorph]

Thanks for all the ideas guys,

an SSTO Rocket Tanker sounds very interesting. Would save me a lot of money. interesting design challenge, too.

Spaceplane tankers haven't worked for me because of the stock soup-o-sphere.

Using multistage rockets though it looks like what accountants call "a wash" it costs basically the same no matter how you do it. (your lifting the same amount of rocket and the same amount of fuel either way)

Off to the VAB!

Edited January 21, 2015 by Brainlord Mesomorph

[GoSlash27]

Not sure I agree with the idea that larger launchers cost more due to expensive engines. You can always cluster the engines used in smaller rockets.

RIC,

Aye, but the act of clustering requires more parts, which raises the cost. It also reduces the structural integrity of the vehicle, which requires more parts to add rigidity and stability.

It should be a wash theoretically, but in practice (at least my practice) it doesn't actually work out that way.

Best,

-Slashy

[Laie]

an SSTO Rocket Tanker sounds very interesting. Would save me a lot of money. interesting design challenge, too.

Spaceplane tankers haven't worked for me because of the stock soup-o-sphere.

Once you've grown accustomed to what jet power can do, the payload fraction of an ordinary rocket SSTO appears abysmal.

Recently, I've been exploring the approach of a recoverable first stage that kicks the whole vessel to a high apoapsis; the next stage then pretty much points towards the horizon immediately, effectively starting it's circularization. When done right, you can switch back to the first stage in time to watch it re-enter and splash down.

The fun part is to make the recoverable stage as weak as possible: the more oomph it has, the farther away from KSP it will end up, depressing the recovery value. The best I could manage so far is ~3200m/s at high TWR with a steep ascent, burns until ~40km with a 60km+ apoapsis. It will fall close enough to still recover at 90%. The next stage needs a TWR of at least 0.6 to get the job done in time; TWR 0.8-1 is preferable.

Another interesting problem is to make sure that the late stage doesn't fly off towards the Mun; when the AP at the start of the burn already is 60km+, it takes some skill to not end up with a 150km apoapsis after all is said and done.

The third problem is that the first stage will SPLASH DOWN in the DEADLY WATER. The only engines that let me do that without mounting a preposterous amount of parachutes are the liquid fuel boosters.

But, back to topic: it's easy to design an efficient tanker. If you use it regularly, tuning and refining will happen as a matter of course. It stands to reason that fuel delivered by such a vessel will be comparatively cheap.

[cantab]

The third problem is that the first stage will SPLASH DOWN in the DEADLY WATER.

Regarding this issue, you can pitch in a direction other than east you know

[Pecan]

Once you've grown accustomed to what jet power can do, the payload fraction of an ordinary rocket SSTO appears abysmal.

This is the truest thing ever in KSP :-) Jets are incredibly OP for efficiency.

Recently, I've been exploring the approach of a recoverable first stage...The fun part is to make the recoverable stage as weak as possible...

Might as well make the whole launch vehicle SSTO, switching jet to rocket when required, detach the payload in orbit and recover the launch vehicle 'at KSC' for 98%, even if you can't get the extra 2% for runway-landing a spaceplane gives you. A VTVL, 'tail-sitter' jet-launched rocket is almost as easy to build and fly as a pure rocket, almost as easy to land as a spaceplane, best of both worlds if you do it right. Having said that, I'm still messing-around with pure rocket SSTO launch vehicles because they are so easy.

[SaturnV]

It's not much about cost efficiency, it's more about ÃŽâ€v potential.

Refueling your ship on orbit can saves you about 4500m/s, thus you can develop a smaller ship for long journey, reduce the difficulty launching the whole thing from sea level at once.

What's more important, there are always dead weight when you addin fuels to your ship (tank itself, support components, etc.)

Say fuel ratio = fuel : dead weight.

In current game the highest value were achieved by X series (8:1).

Then there's a ÃŽâ€v limit for a single stage, and the weight of vessel increase significantly for multi-stage rocket. So you want to built a heavy rocket with low ÃŽâ€v: Sent the ship to orbit with empty tank (~4500 dv), then push tons of fuel to orbit (~4500 dv).

That will be easier than orbit in one shot for heavy payload. Besides, it helps reduce the part count

[Laie]

[recoverable first stage]

Might as well make the whole launch vehicle SSTO,

For an SSTO, efficiency is everything, and using overpriced parts for the sake of a small performance gain is justified. Fully recovering a 500k vessel is cheaper than throwing away a 200k one, after all. I started designing my first rocket SSTO tanker when Mk3 parts came out. That rocket was a real beauty, sleek, art deco, reminiscent of a 1930s high-speed railroad. Also fiendishly expensive, to the degree that give or take 2% recovery value started bothering me. Then I thought, this is getting silly.

The splashed-down 1st stage recovers for a "mere" 90%, but that's 90% of fewer and cheaper parts. It *is* more expensive overall (98% is hard to beat), but not by as much as you may expect. Side benefits: Launches are FAST! Have to be, of course, because you need to be done before the first stage drops below the threshold. Don't know about you, but my SSTO rockets still took about 10-12 minutes to orbit, and then just as long to return them to KSC. The 1st-stage approach needs six minutes to orbit and no de-orbit burn; what's more, most of the reentry is over by the time I get to switch vessels.

For tankers, I returned to spaceplanes. Sending an overpriced and -sized rocket all the way to orbit just because I can't recover it otherwise feels silly; with spaceplanes I have no such qualms, even though they're just as expensive. Besides, if I have to de-orbit and reenter, I may as well aim for the runway and do it in style. But everything that is supposed to remain in space rides up on a recoverable first stage. And if I had to use a rocket tanker, it would also make use of the concept, if only because it lets me launch more fuel for the part count, and most of the parts are gone after two minutes.

TL;DR: I know why you prefer rocket SSTOs over spaceplanes. The reasons why I prefer a recoverable first stage over a rocket SSTO are pretty similar. I think you should give it a try.

Edited January 22, 2015 by Laie

[Aristarchus]

I am a big fan of orbital refueling stations. While it may be less efficient overall for single launch missions, you can quickly recover the extra cost by reusing some vehicles for later missions. For example, I have orbital stations around Kerbin, Minmus and the Mun. I send an SSTO crew shuttle to my LKO station where I have a shuttle consisting of a capsule, hitchhiker container, fuel tank and nuke engine, which takes my crew to Minmus. There a lander is waiting for them to complete a contract or just land for experience. Afterward they can either transfer to the Mun or head back to Kerbin LKO where they wait for a new crew to relieve them. I can do dozens of missions in Kerbin SoI before I even have to think about refueling, all without throwing any new hardware into space and thus saving countless booster stages.

Edited January 22, 2015 by Aristarchus

[Brainlord Mesomorph]

On the subject of an SSTO Rocket Tanker: it seemed silly to carry around the giant terrestrial rocket engines and empty fuel tanks on orbital maneuvers. So I came up with this:

That is my standard orbital Fuel Tender nestled comfortably between two of my brand new “Saturn SSTO†booster rockets. Each one apparently has a payload capacity of about 35 tons, and between them they can easily lift this 70 ton payload to orbit, and still have exactly the three drops of fuel needed for a deorbit burn.

Interestingly, the two of them only cost a fraction more than my previous “70 Ton Lifter†multistage rocket that was entirely disposable. (Maybe I could recover a big gray tank and a mainsail from the top stage) these are 100% recoverable, and don’t even leave debris in orbit!

Even if I splash these down on the far side of Kerbin for 50% recovery value, it still costs half of my previous launch system. And it should be scalable, if 2 can lift 70 tons, four should be able to lift 150, and eight should be able to carry 300 tons!

So again, thanks guys, this is a more efficient solution.

UPDATE: Those things are going to need WAY moar parachutes and perhaps some landing gear I am going to recover them!

Edited January 22, 2015 by Brainlord Mesomorph

[Pecan]

..So I came up with this...

Now that's the sort of stuff I'm talking about, although more complex than I'd do it. Yes, it's scalable; a KR-2L can SSTO 26t, so three of them 78t, with enough fuel for de-orbit and 'chute-assisted powered landing. I'd suggest trying drogues more than 'chutes, letting them bring the descent-rate to 20-50m/s, which leaves hardly anything for the engines to do to touchdown safely.

As I posted elsewhere, here's SSTO 40 parking at the office. It uses mainsails (SSTO 11t each) and delivers a 40t+ payload, e.g.; orange tube + accessories, to orbit.

10 orange tubes to launch one is very fuel-hungry even by rocket standards, let alone jets but it's only 42 parts including batteries, etc. and flies a normal ascent and landing. 98% recovery 'at KSC' is good enough for me.

... The reasons why I prefer a recoverable first stage over a rocket SSTO are pretty similar. I think you should give it a try.

You are almost certainly right, if I weren't too busy already I would be trying it :-) At the moment I'm still thinking it would be worth dragging the jets the rest of the way to orbit rather than mess about with recovering two stages but I can't comment since I have no data. 'Worth' is pretty time/effort subjective anyway in this case so it's just a preference call. NB: I'm not saying SSTO rockets are more efficient, they definitely use a lot more fuel, I'm just saying that I use them at the moment because they're easy and low part-count.

Edited January 22, 2015 by Pecan

[Laie]

10 orange tubes to launch one

[...]

At the moment I'm still thinking it would be worth dragging the jets the rest of the way to orbit

Jets? I was talking about a 100% rocket solution. In the case of turbojets, you need to fly a shallow ascent in order to milk them for their thrust; by the time you're done with them, they would end up on the far side of the world for a 50-60% recovery value. I can't imagine that keeping them with you will turn out to be more expensive.

*opens his vessel and picks up calculator*

590t of fuel on the pad, >100t left to spare in orbit. If I swapped the tanks for another payload, it could weigh ~110t. That's more like, what, 5:1? Not carrying that first stage into orbit really saves a lot of fuel. Recovery losses are about 30k... *types* ...two-stage is slightly cheaper than spending twice the fuel and recovering at 100%.

[Pecan]

Jets? I was talking about a 100% rocket solution...

Ahh, I misunderstood because "Once you've grown accustomed to what jet power can do, the payload fraction of an ordinary rocket SSTO appears abysmal. Recently, I've been exploring the approach of a recoverable first stage..." in your earlier post. From that I'd assumed you were using a recoverable jet first-stage but just hitting 'up' as hard as you could insted of milking it.

To be clear - what payload mass are we talking about here, I'll try to play-around with different approaches? (Away for the weekend though)

[magnemoe]

Depends on what kind of SSTO it is. If it's a spaceplane I think refueling in orbit is cheaper but if it's an SSTO rocket then it probably doesn't matter too much and it would be easier to do in one launch.

Yes an spaceplane SSTO use little fuel but has size constraints, SSTO rockets can be build as large as you want, I made an 200 ton cargo to LKO once so here its better to refuel.

Still I would not use an spaceplane for an mun or minmus mission. An dedicated craft with LV-N engines would be cheaper and easier to do, you can still refuel it from the SSTO.

I use an 10 man plane and a 10 man spaceship / lander who do the qualification runs on 720 liter fuel. Mun orbit transfer to Minmus for landing, then outside Kerbin SOI and back to LKO.

If you want to land on Mun here I recommend bringing a service module with spare fuel for the rest of the trip, leave this in orbit and do the mun landing with 2km/s then pick up the fuel before you continue.

[Ruedii]

I forgot to mention the issue of necessary thrust to weight.

Once you are in orbit, you can use a low thrust to weight (ranging from 0.1 to 0.8 depending on how you like) but your initial launch requires high thrust to weigh.

It's a lot easier to launch fuel in a craft designed to only carry fuel to orbit, and launch the craft with it's tanks mostly empty, in order to get the necessary thrust to weight off the ground without resorting to heavy rockets like mainsails.

[Laie]

To be clear - what payload mass are we talking about here, I'll try to play-around with different approaches? (Away for the weekend though)

Any, really. The basic building block I use the most is one LFB plus one-and-a-half orange tanks. If arranged in a shape that won't tip over in the water, this will survive splashdowns at reckless speed. Also, TWR can't possibly be too high. Slap on as many of these as necessary to give you TWR~1.6 / 3200m/s to start with, aim for an apoapsis of about 60-70km by the time this burns out. In terms of orbital velocity, this will only provide 500m/s or so (that's why it can still be recovered close to KSC), but the next stage can already look like a proper orbital vessel, with TWR<1 and space-faring engines.

If your payload happens to be a proper space vessel, it may well be able to circularize under it's own power, or only needs to be pimped up with a few extra engines that really aren't worth recovering.

Edited January 22, 2015 by Laie