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

[Brainlord Mesomorph]

I have a vehicle that is fully capable of SSTO Kerbin. But of course, it uses all of its fuel to get there. And that’s not its mission. Its mission is interplanetary, and it must leave LKO fully fueled.

So the question becomes what is cheaper: launching the vehicle under its own power and refueling in orbit? Or putting it on the top of a rocket, and hauling it to orbit fully fueled?

I can see advantages both ways. It seems to me there should be an “economy of scale†for delivering large amounts of fuel to orbit and refueling several ships. On the other hand, given rocket staging, is might be cheaper to send several smaller payloads.

Anyone have any insight? Any opinions?

Anyone with any actual math?

As always,

AHA , TIA

Brainlord

Edited January 21, 2015 by Brainlord Mesomorph
answerred

[Alshain]

In theory and if done correctly, launching at once should be more cost efficient. However, refueling in orbit is more fun

This is unless you have an infrastructure for refueling. You can build a refueling SSTO would would be significantly more cost effective and fun. Two launches, full recovery. Fuel is cheaper than rocket engines.

Edited January 20, 2015 by Alshain

[Red Iron Crown]

Single launch is more efficient.

Payload fraction doesn't change with payload mass; e.g. if you can put 10 tons up with a 60 ton rocket you should be able to put 100 tons up with a 600 ton rocket. Furthermore, once in orbit using multiple launches some propellant must be spent to rendezvous and dock, plus you need the additional mass of docking ports.

That said, if you find it more fun to assemble or refuel in orbit you should do it as it's not a huge efficiency difference, generally speaking.

Edited January 20, 2015 by Red Iron Crown

[Frank327]

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.

[RainDreamer]

IMO, the only thing that should prevent you from launching everything at once is when it is simply impossible to take off into orbit. Otherwise, haul as much as you can to space in one launch.

[Brainlord Mesomorph]

In theory and if done correctly, launching at once should be more cost efficient. However, refueling in orbit is more fun

This is unless you have an infrastructure for refueling. You can build a refueling SSTO would would be significantly more cost effective and fun.

LOL I have more than enough opportunities for docking.

My question is purely economic.

infrastructure for refueling: My experience is that Orbital Gas Stations were a waste of money (all the fuel wasted trying to dock huge amounts of fuel). I do better with single use Fuel Tenders that go up, refuel a ship or two and then are deorbited and recovered.

I tried a fuel-carrying spaceplane, It always burned more than half the fuel before it got there. (and took forever!)

- - - Updated - - -

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.

I have both. The one in question is an overbuilt lander.

But would wings mattter? SSTO is SSTO.

[cantab]

If all your lifters are balanced in terms of cost per ton to orbit it shouldn't make much difference. Having the ship contribute to its own launch will let you use its engines though. A mixed approach might be good - strap some boosters to the ship so it can make orbit without using all its fuel, then send a tanker to top it up.

If you have vehicles with widely varying cost per kilo, then you want to use the cheapest. If the cheapest isn't capable of lofting the ship full - for example it's a spaceplane - then you need to refuel.

If you're playing without reverts you also need to consider development costs. Large lifters can be troublesome, there's a lot to go wrong, and it's a massive cost if a launch fails. Sending a series of fuelling missions with a tried-and-tested lifter or a simple development of it is a much safer option.

And don't forget to factor in the IRL length of the mission. 20 spaceplane flights to refuel might not be much fun. Then again one launch taking an hour at seconds-per-frame game performance might not be much fun either.

[Wanderfound]

1) Launch interplanetary ship on minimal boosters, using all of its fuel to get to orbit.

2) Launch spaceplane tanker, refuel interplanetary ship.

3) Land tanker at KSC for 100% recovery minus fuel.

4) Profit.

This is much easier in FAR than stock, however. Large spaceplanes are painful to fly in stock air.

[Brainlord Mesomorph]

1) Large spaceplanes are painful to fly in stock air.

oh, THAT explains it.

[WuphonsReach]

And the calculations change if you have a mining / refueling infrastructure built on Minmus (with MKS/OKS)

[Alshain]

This is much easier in FAR than stock, however. LargeAny spaceplanes areis painful to fly in stock air.

Sorry, I just had to fix that

[cantab]

And the calculations change if you have a mining / refueling infrastructure built on Minmus (with MKS/OKS)

Indeed. if you can ship fuel in from Minmus - or almost anywhere that's not Kerbin really - that can offer significant savings.

[Streetwind]

Even when not using spaceplanes, launching fuel separately could be potentially interesting. Very large ships require very expensive, large diameter liquid rocket launch stages which are expensive and difficult to recover properly. Small payloads, however, can be kicked into orbit by all-solid boosters and a bit of RCS (or a liquid engine feeding from the tank's own fuel) to circularize.

I am excessively using ICBM-style solid rockets in my current game to launch RemoteTech stallites, for instance. One RT-10 trashcan costing 325, a decoupler for 200, a solid upper stage for 220 (from RLA Stockalike, but it probably would work with a second RT-10 too), another decoupler for 200. With FAR installed, this is easily enough. So: less than 1000 funds for a (roughly) 1-ton payload, requires almost no tech unlocks, and I can launch three of those in the time it would take a spaceplane to fly back and land at KSC for recovery... I'll gladly pay that bit of chump change for expedient commsat delivery. And since the solid boosters decouple while the periapsis is still in the atmosphere, I'm not even littering my orbit. I call it a "small dumb booster approach to economic launches"

Using the bigger solid boosters, I'm sure you could send at least a FL-T800 to the point where only a couple dozen m/s remain for orbit insertion, even under stock aero, for a few thousand a pop. You could probably launch ten of those for the price of the extra liquid boost stage that it would take to get a fully fueled Jool spacecraft into orbit instead of launching it empty. Maybe it's not price competitive with spaceplanes, but it's quicker, and it should still be price competitive with a fully fueled launch strategy.

[eddiew]

This is unless you have an infrastructure for refueling. You can build a refueling SSTO would would be significantly more cost effective and fun. Two launches, full recovery. Fuel is cheaper than rocket engines.

http://i.imgur.com/IFVfgoU.png

That's not a tanker, it's a hotrod

[Alshain]

That's not a tanker, it's a hotrod

Why can't it be both?

[flashcactus]

Well, the statement about fuel being cheaper than rocket engines is very true. Unless the stage you're hauling the SSTO up with is fully returnable or dirt cheap (read: entirely or almost entirely consisting of SRBs), using a returnable fuel tanker is significantly cheaper money-wise. It *is* more expensive part-count-wise and mass-wise, though. Also, if you use something like RealFuels, maintaining an orbital refueling depot suddenly becomes a significantly harder task.

[Pecan]

1) Launch interplanetary ship on minimal boosters, using all of its fuel to get to orbit.

2) Launch spaceplane tanker, refuel interplanetary ship.

3) Land tanker at KSC for 100% recovery minus fuel.

4) Profit.

This is much easier in FAR than stock, however. Large spaceplanes are painful to fly in stock air.

In stock:

2) Launch SSTO rocket tanker, refuel interplanetary ship.

3) Land tanker at KSC for 98% recovery minus fuel.

This IS easy.

[Kesa]

The advantages I can see are rather real-game-time related than kerbal price related.

Bigger ships -> fewer launches, fewer docking (if any)

Smaller ships -> simpler design (and faster to do), more stable, easier to fly

Plus a refueling vehicle is only designed once for any mission involving ship of similar or (reasonably) greater size.

For automated refueling, you should try this.

In kerbin, the cost per pound decreases as the playload mass increases because bigger engines have better TWR.

But jet powered spaceplane SSTOs are the most cost effective, so if only cost efficiency matters, you should use them for anything more than few tons (as long as part count/lag allows you). Instability of big ships can induce steering loss, and docking small ones require slighty more dV (loss is induced for all but first launch, so there is more difference 2 and 1 launches than between 10 and 5), but this is very small.

If you overpower your spaceplane or pack one or two ditchable SRBs, you can fly it like a rocket, saving time at launch, and can glide to runway after reentry.

The real life argument doesn't really apply to KSP :

Bigger is always better, as long as structural feasability is met (unless using mods, adding more struts makes this less of an issue in ksp), due to the square cube law, which reduce drag (exists with far) and tank dry ratio for big rockets (exists with procedural tanks from Real Fuels or procedural part, I can't tell).

However, generic launchers are far cheaper than designing a new bigger one (designing is a significant part of a rocket), even though a new bigger one can be useful if use for other missions (exists, to a limited extent, with rp 0, with part cost quite low but unlock cost quite high)

- - - Updated - - -

.

I second the Idea of SRBs, but I have to add that given enough of them, you could also go for a fully fueled launch strategy.

Edited January 21, 2015 by Kesa

[WuphonsReach]

I don't know about you, but my refueling pods are all of identical design. Enough DV at liftoff to get them into 85km orbits without burning much of the cargo, with a docking port on the front and minimal extras.

If a particular vessel doesn't need all of the fuel, it doesn't have to take it all. If a particular vessel needs multiple pods to fill it up, that's okay as well.

[GoSlash27]

I'd say it's cheaper to refuel in orbit.

The larger engines required for larger launch vehicles are disproportionately expensive.

Plus larger vehicles are less structurally rigid and thus require more struts and are more difficult to operate safely and recover intact.

Best,

-Slashy

[tomf]

If you are using staged launch vessels and not expecting to recover the lower stages then it is going to be cheaper to do multiple launches as the cost of larger lower stage engines goes up a lot faster than their thrust. I guess if everything is recovered for a large part of its launch cost then you are going to save yourself considerable headache trying to launch two smaller SSTOs rather than build one huge one.

[Red Iron Crown]

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.

[Toastedbuns]

Single launch is more efficient, but having an orbit station with fuel for interplanetary journeys and the like is actually much more convenient I find once you get docking down. I recently did my first dock (a couple days) and since then I've been designing a very low orbit (75x75) fuel station to refuel any spent fuel of the interplanetary stage so the journey is a guaranteed success. Cause correcting delta v needlessly is a complete waste of time, right? Why fix my mistakes when I can just siphon fuel from a hunk of metal I launched into space for fun?

[Superfluous J]

Why fix my mistakes when I can just siphon fuel from a hunk of metal I launched into space for fun?

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

[michaelsteele3]

*Insert lazy voice here*

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

Ew. Those words make me sick.