Why doesn't Nasa, Space X etc use heavy lift rockets that carry multiple payloads into orbit?

[TeeGee]

For example:

Instead of using 1 rocket to launch 1 satellite, use a heavy lift rocket that will carry 4-5 or more satellites, a cargo module for the ISS AND maybe even a crew module into orbit.

I don't understand how it could be more beneficial for rocket companies to have multiple small payload launches as opposed to 1 or 2 multiple payload launches using heavy lift systems.

In KSP, this is what I've been doing. Instead of launching Kerbals, satellites and supplies on separate rockets; I put them all onto 1 big one. I launch the Kerbals in LKO, move into a higher orbit and launch my satellites, then move the SM into a graveyard orbit. I just delivered 5 launches worth of payload into 1.

Wouldn't that save money and time?

[radonek]

Eggs and baskets sir, eggs and baskets.

[AngelLestat]

they launch multiple payloads some times, but is not as easy to arrange that.

All satellites needs different orbits, sometimes there is not a chance to wait 3 years until another client wants to launch a "right side" satellite in a similar orbit.

[Jovus]

Plus (at this point) R&D costs for new rockets, plus (though I doubt doubt this is a smaller factor) for some mission profiles it wouldn't work very well, and you might not save delta-V because of extensive orbital maneuvering.

As a side point off the top of my head, I wouldn't be too surprised if the $/dV were cheaper in the first stages than in the latter stages, since RP-1 and storage is probably cheapter than hydrolox or hypergolics.

That's all just searching for plausibilities, though. I can't say I know the answer, because I don't work in the industry (yet).

[NovaSilisko]

Don't use KSP as a justification for why something should be done in real life.

That many payloads just become a ludicrous pain to coordinate. If your rocket is set up to launch 6 major payloads, and then, oops, 5 days before the launch, one of the heaviest ones runs into a major technical flaw and needs another month, you'll have to scrub the launch, and hold up every other payload on the vehicle, because otherwise it would be delayed even more to transfer them to other launchers, or to reconfigure the launcher to fly with only 5 payloads on board. At which point, the 6th is left high and dry, waiting potentially years for another ride-share opportunity on another large rocket.

The same applies in reverse as well. If the rocket encounters an issue shortly before launch day, now 6 payloads are out of luck until it's fixed. Moving payloads from one rocket to another is not trivial or easy by any standard.

[moronwrocket]

There are plenty of times that multiple satellites are launched with one payload. Herscel and Plank were launched together, for example. Many launch vehicles offer standardized secondary payloads, with Cubesat deployment modules a common secondary payload. http://en.wikipedia.org/wiki/Ariane_5 for example talks about Ariane 5 supporting up to 8 secondary payloads.

Why are they not done more often?

Well, a manned launch is going to be much more stringent on safety, and probably wouldn't want to deal with complex secondary payloads. Also, insurance is bad enough with one payload carrying rocket fuel, imagine if you have a bunch of other payloads that also have rocket fuel to reach their own final orbits to deal with!

Plus, the flight profiles need to match up - paying for a plane change is expensive, or you might not be able to find a rocket-roomy interested in the same destination you are heading.

Mission timing could also be a factor - with 5 paying payloads, there's 5x the chance something delays one - do you stop the whole launch for one tardy satellite?

Beyond EC, KSP doesn't simulate the hassles of a real multi-vehicle launch - crews checking out their payloads, comm systems, etc. That's probably a factor as well.

[cpast]

Ariane already does this routinely with 2 satellites, and it's not uncommon to have a number of CubeSats on another launch because they're so small. However, there are issues with launching tons of things on one rocket. First, it only works if they're all going into mostly similar orbits -- if you have them going into different orbits, then you need to waste a lot of fuel moving things around between orbits (GTO is an exception, as it's not rare to have two payloads going to GTO). Keep in mind that real orbits are mostly not equatorial and at similar altitudes; real orbits are almost always at some inclination, often a significant one (again, GEO is an exception).

Also, it's highly questionable whether you'd be saving money. The issue with a huge rocket is that they're really, really, really, really expensive; you need a *lot* of payload to make that worthwhile. Unless you have tons of missions that happen to be ready at around the same time, and which all take similar orbits and all fit on the rocket, you'll end up substantially under your weight budget, which means the launch will be very expensive compared to a smaller rocket. It's similar to why airlines fly regional jets instead of using 747s everywhere, or why mail carriers don't use semis on their routes -- having a huge capacity is only sensible if you can actually fill that capacity in a reasonable amount of time. In the actual launch market, there just aren't enough payloads to be able to reliably fill a huge rocket going to a specific orbit with 5-6 of them; that means a customer has to either wait a long time for other customers who can share the launch with them, or pay the cost of a massive rocket without many customers to split it between.

Combining a *crew* module with cargo is what happened on the Space Shuttle, and was one reason it was so expensive. Manned spacecraft have to meet extremely high standards for reliability, and you don't need that for cargo (you can get by with more risky launches and an insured payload). Manned missions are also *really* expensive; combining with cargo requires the cargo customers to subsidize the manned mission to some degree. One lesson from the Shuttle is that building a craft to take a crew *and* cargo means the cargo launches end up being expensive.

[cantab]

Orbital constraints and general practicality. Especially as unlike in KSP, real rocket engines can only ignite a finite number of times, often only once.

Ariane 5 is one example of a rocket built with launching two payloads at once in mind. It's caused problems. Satellites have tended to become heavier than they used to, and the Ariane 5 is too expensive to economically launch a single satellite so Arianespace needs to find a "light" satellite to pair with a "heavy" satellite, which causes delays. Overall Ariane 5 is reliable and reasonably cost effective, but it's not overwhelmingly cheaper than other one-payload launchers.

[Beowolf]

SpaceX has mentioned this as a goal for Falcon Heavy. They said their lower cost should overcome customer resistance due to those very real problems Nova and others above have already described. Plus they're planning to launch something like one a month, which could make rescheduling easier, depending on specific orbits and such.

[PlonioFludrasco]

Last year, Russians launched 32 satellites at once, atop a Dnepr rocket. (demilitarized version of the "Satan" ICBM, and still launched from an underground silo!!)

[Dunbaratu]

Doesn't the size of the rocket scale in a way that isn't linear with the size of the payload?

In other words, to put twice as much mass in orbit requires MORE than twice as much of a rocket.

It seems that this would make two smaller launches of half the payload each more economical than the one combined launch.

[rkman]

Doesn't the size of the rocket scale in a way that isn't linear with the size of the payload?

In other words, to put twice as much mass in orbit requires MORE than twice as much of a rocket.

In theory (only looking at the rocket equation) it does not. But in practice, bigger heavier things need to be more sturdy to be able to carry their own mass.

It is why for a given material there is a maximum hight of a column build out of that material (with a given diameter and gravity), and it is why the legs of large animals are relatively thick.

Otoh i don't think it scales linearly, and i'm not sure at what point it becomes a significant factor in rocket construction.

[GoSlash27]

Teegee,

I doubt you're saving any money doing it that way even in KSP. Economy of scale doesn't really work in space and it's wasteful to use more rocket than you need for a payload.

Time, sure... so long as you happen to have multiple payloads that can go in the same launch window. Due to the limitations mentioned upstream and more, this rarely happens IRL. For example, we don't launch stuff from the equator directly into a zero inclination orbit IRL. We tend to do this in KSP, which makes windows a lot less demanding.

Best,

-Slashy

- - - Updated - - -

In theory (only looking at the rocket equation) it does not. But in practice, bigger heavier things need to be more sturdy to be able to carry their own mass.

It is why for a given material there is a maximum hight of a column build out of that material (with a given diameter and gravity), and it is why the legs of large animals are relatively thick.

Otoh i don't think it scales linearly, and i'm not sure at what point it becomes a significant factor in rocket construction.

As a practical matter it does, but the important point is that it definitely doesn't improve with size. Therefore there is no economy of scale to be had.

Best,

-Slashy

[Silver_Swift]

In theory (only looking at the rocket equation) it does not. But in practice, bigger heavier things need to be more sturdy to be able to carry their own mass.

It is why for a given material there is a maximum hight of a column build out of that material (with a given diameter and gravity), and it is why the legs of large animals are relatively thick.

Otoh i don't think it scales linearly, and i'm not sure at what point it becomes a significant factor in rocket construction.

I don't think the square cube law applies here. I mean, worst case scenario: you duct-tape a bunch of smaller rockets together and it will be as structurally sound as the individual rockets were (since each rocket is supported by its own thrust).

The square cube law only applies when the supportive structures scale with surface area whereas the stuff that needs supporting scales with volume. An animal that is twice as big in every direction weighs eight times as much while bone cross section is only four times as big, but if your rocket has a payload that is twice as big in every direction it will carry eight times as much payload and it will need eight times as much thrust.

You can't have a elephant sized mouse, but you could totally have a elephant size pack of mice each supported by their own legs.

Edit: swapped mouse and elephant. Thanks smjjames .

Edited February 25, 2015 by Silver_Swift

[RainDreamer]

Didn't JAXA launched a bunch of satellites at once for their Hayabusa 2 launch? I think it is just depending on how big the satellites are, and whether they are supposed to be in those orbits together.

[YNM]

If you get their intended orbit and mass right, it's easy to get one. But imagine launching an imaging sattelite (so sun-synchronous orbit) at once with a GTO commsat, which I think both are among the most common ones...

I guess that until we start to miniaturize most things apart from those directly constrained by physical limits (like comm dish, telescope, solar panels etc.) we won't be seeing these options come to fruition. And also, a too efficient upper stage.

Edited February 25, 2015 by YNM

[smjjames]

You can't have a mouse sized elephant, but you could totally have a elephant size pack of mice each supported by their own legs.

I think you meant elephant sized mouse because you could totally have a mouse sized elephant.

[TeeGee]

But how about the economics of it all?

Calculate how much an average satellite launch will cost per launch on a single payload rocket. Then calculate how much a delta-4 heavy rocket launch will cost. I'm sure there is a point in the delta 4 heavy payload where it would be cheaper to launch multiple satellites atop the delta 4 heavy than to use multiple single payload rockets.

And I don't understand why the amount of delta v used wouldn't be beneficial. Heavy lift rockets are designed to push higher mass objects into LEO. Arrange payloads delivered top to bottom according to orbital altitude, decouple them, then proceed to the next orbit needed. The SM would be shedding weight as it went up as it delivered the satellite payloads.

Delta 4 heavy launch with max payload weight worth of satellites on top. Delta 4 H has a max payload to LEO of 63,470 lb, 31,350 lb to GTO... that's a pretty hefty payload. A modern satellite can weigh UP TO 6 tons. That's 5 satellites in one go.

How much are 5 separate satellite rocket launches vs. 1 delta 4 H launch?

Delta 4 heavy launches cost $435 million each. Space X falcon 9 launch is $62.5 million.

62.5 x 5= 312.5 million. So its cheaper to do multiple launches with space x than one big one.

BUT the average satellite launch for everyone else is upwards of $100-250 million PER launch. 100 x 5 = $500 million.

And as a satellite company, I wouldn't be paying for the whole damn delta 4 rocket... I'd be paying for my contributing payload mass. If 5 different companies each want to launch 1 satellite up on a delta 4 heavy, they'd be splitting the $435 million 5 ways. That's $85 million each. CHEAPER than the average.

Like it doesn't make sense to me why they don't do that. Fewer small launches means LESS orbital debris (a huge problem these days) and it would be cheaper for the company hiring to use the rocket.

Edited February 25, 2015 by TeeGee

[Kryten]

Nobody is putting constellations of large satellites into LEO, and LEO payload figures are in most cases a red herring. Delta IVH can put 14 tons into GSO, which would be just under 3 large modern comsats. The go-to launcher for putting something about 5 tons GTO alone would be proton, at about $70 million per launch; whereas Delta IVH is somewhere in the region of $400 million.

Also, that's not how pounds and tons work.

[YNM]

Depends on which ton. Metric ton are 1000 kg... long and short ton are there too, with even sillier conversion. But I guess most people already know that metric ton is preferred ? (seriously, I don't want units war, tell me if I'm wrong. But I guess most scientific community should adept SI units)

And as I said, not all launches are going to GEO/GTO. SSO and some other inclined orbit (like Molniya) also frequently used. Not to count missions to other bodies or Lagrange points, ISS, custom orbits that copes for the satellite's job (positioning system sats, weather sat for different nations, deep space imaging sat like Hubble, Keppler, Gaia, Fermi)...

[RainDreamer]

I think there is also the problem of just packing things into the available payload space, then release them in a way that doesn't smack things together and destroy everything else.

[GoSlash27]

But how about the economics of it all?

Calculate how much an average satellite launch will cost per launch on a single payload rocket. Then calculate how much a delta-4 heavy rocket launch will cost. I'm sure there is a point in the delta 4 heavy payload where it would be cheaper to launch multiple satellites atop the delta 4 heavy than to use multiple single payload rockets.

And I don't understand why the amount of delta v used wouldn't be beneficial. Heavy lift rockets are designed to push higher mass objects into LEO. Arrange payloads delivered top to bottom according to orbital altitude, decouple them, then proceed to the next orbit needed. The SM would be shedding weight as it went up as it delivered the satellite payloads.

Delta 4 heavy launch with max payload weight worth of satellites on top. Delta 4 H has a max payload to LEO of 63,470 lb, 31,350 lb to GTO... that's a pretty hefty payload. A modern satellite can weigh UP TO 6 tons. That's 5 satellites in one go.

How much are 5 separate satellite rocket launches vs. 1 delta 4 H launch?

Delta 4 heavy launches cost $435 million each. Space X falcon 9 launch is $62.5 million.

62.5 x 5= 312.5 million. So its cheaper to do multiple launches with space x than one big one.

BUT the average satellite launch for everyone else is upwards of $100-250 million PER launch. 100 x 5 = $500 million.

And as a satellite company, I wouldn't be paying for the whole damn delta 4 rocket... I'd be paying for my contributing payload mass. If 5 different companies each want to launch 1 satellite up on a delta 4 heavy, they'd be splitting the $435 million 5 ways. That's $85 million each. CHEAPER than the average.

The way to look at it is $/tonne rather than $/launch. Moreover, each satellite going to a higher/ different orbit weighs a lot more than just the payload; you need the rocket/ fuel/guidance/ structure to place the payload there.

Like it doesn't make sense to me why they don't do that. Fewer small launches means LESS orbital debris (a huge problem these days) and it would be cheaper for the company hiring to use the rocket.

Not to sound dismissive, but it makes sense to the folks who make such decisions. If they were saving money doing it that way, they would. And it wouldn't cut down on orbital debris. In fact, it would add to it. Remember that everything that's not circularized comes back down.

Best,

-Slashy

[rkman]

I don't think the square cube law applies here. I mean, worst case scenario: you duct-tape a bunch of smaller rockets together and it will be as structurally sound as the individual rockets were (since each rocket is supported by its own thrust).

Realistically there is a limit to how wide a rocket can be made before that becomes a problem re aerodynamics.

So generally, rather than duct-taping rockets into a bundle, a rocket would become wider and taller as it becomes larger, and then the square cube rule applies.

[Silver_Swift]

I think you meant elephant sized mouse because you could totally have a mouse sized elephant.

Ah, crap, you're right. I thought I checked that I had them the correct .

Though, to be pedantic, you probably couldn't have a mouse sized elephant either as I believe there are a few mechanisms for which the square cube law works in reverse (heat management is the one that comes to mind, you lose heat based on surface area and gain heat based on volume).

Realistically there is a limit to how wide a rocket can be made before that becomes a problem re aerodynamics.

So generally, rather than duct-taping rockets into a bundle, a rocket would become wider and taller as it becomes larger, and then the square cube rule applies.

I'm having trouble picturing this in my mind, if one rocket can get to space on its own, why can't multiple rockets that are effectively just flying very close together?

Of course, you might be able get some aerodynamic advantage out of the fact that you are launching all of them at once by stacking them on top of each other, but that works in favor of one large rocket rather than against it.

Edited February 25, 2015 by Silver_Swift

[Kryten]

Trying to get multiple large payloads for any deep space destination at current demand is an exercise in futility, and with the current state of the SSO/LEO market it would be pretty close. D-IVH isn't going to be able to put up station resupply missions this way due to waiting for other payloads inevitably messing up the scheduling, Russian or Chinese government sats aren't going on one before hell freezes over. Launch of NGSO communications satellites like Iridium or Globalstar are also non-starters; it wouldn't be possible to launch more than one plane at a time, and even for the large Iridium sats a single plane is well within the capability of Proton.

That leaves you with the commercial remote sensing market, which isn't much; the average mass of a satellite in this class is well south of a ton, and there are only a few launches a year even with current launchers. You'd have to wait decades to fill something like a D IVH.