Tankers. A new price category?

[charliem]

I have an idea and a question about its feasibility. I'm no space engineer, nor knowledgeable enough in this field to develop an answer, so I'd like to ask it here.

In a mission to the Moon about 3/4 of the mass lifted to LEO is fuel. Going to GEO the ratio is not that much better, and going as far a Mars is even worse.

To date we can classify launch vehicles in two categories by price: very expensive (for cargo) and extremely expensive (for humans).

When the cargo value is in the same range as the vehicle (or above) is sensible to make it as reliable as possible, and that implies cost, but lately we've been toying with the idea of fuel depots in orbit, and fuel is cheap, almost dirt cheap (and food, air, and other human consumables are not much more expensive).

Paying $200 M to deliver to LEO 20 mt of cargo which value is less than $200.000 looks plain dumb unless there's no other way.

Is there?

If we could determine a function for rate of success vs cost, and accepting cargo value is negligible against launcher value, should be possible to calculate the sweet spot where we can get the most mass in orbit for the least money.

Complex engines? Triple redundant avionics? 40% structural margins? Reusable hardware? Extensive monitoring? ... How much of that could we rub off?

So, summing up, my question is: How much could the cost of a launch vehicle be taken down if we accepted a significant decrease in its reliability? (how about 80% or 90%, 1 or 2 losses in every 10 launches?)

P.D. This post was inspired by the Mobile Hab for Mars exploration thread: http://forum.nasaspaceflight.com/index.php?PHPSESSID=2a77e4ff1052dcde92a028a415de5e02&topic=17360.msg417425#msg417425

[savuporo]

You probably want to read up on MCD ( minimum cost design ) ideas for space launch.
These tend to go towards ablative nozzles, pressure fed or low-cost pump designs ( piston pumps, flometrics style pistonless pumps) etc.

MicroCosm has been the recent proponent of the concept, with their Scorpius launcher plans.

Of course, they are always two years away from their first launch as many others.

[deltaV]

A malfunctioning fuel tanker rocket puts innocent buildings, satellites and people at risk, not just the fuel. A disabled tanker in orbit could explode from overpressure, scattering debris that put other spacecraft at risk. A malfunctioning navigation system could cause the rocket to shoot for the stars but hit London instead. Liquid oxygen is cheap but a worker killed in a liquid oxygen fueled fire is not. A cheap payload allows one to cut some corners, but many of the most expensive corners can't be cut.

That said a lot of people (myself included) guess that if there were a reliable market for propellant in orbit someone would find a way to provide it cheaper than current launchers.

[Jim]

A malfunctioning fuel tanker rocket puts innocent buildings, satellites and people at risk, not just the fuel. A disabled tanker in orbit could explode from overpressure, scattering debris that put other spacecraft at risk. A malfunctioning navigation system could cause the rocket to shoot for the stars but hit London instead. Liquid oxygen is cheap but a worker killed in a liquid oxygen fueled fire is not. A cheap payload allows one to cut some corners, but many of the most expensive corners can't be cut.

That said a lot of people (myself included) guess that if there were a reliable market for propellant in orbit someone would find a way to provide it cheaper than current launchers.

No, you don't understand the concept.  It is not a cheaper payload, but cheaper launch vehicle.  Also, the cheaper  launch vehicles aren't blowing up left and right.  They are just simpler.

[charliem]

A malfunctioning fuel tanker rocket puts innocent buildings, satellites and people at risk, not just the fuel. ... A cheap payload allows one to cut some corners, but many of the most expensive corners can't be cut.

Alright, some corners can't be cut, but most? ... I bet some could, and wonder how much doing those'd reduce the price tag.

Of course we are not talking about launching from Oklahoma, and of course you can't dispose of the Range Safety System, but what about triple avionics? Why not just one and accept some loss once in a while?

And why a 40% margin on structure, why not 15%?

Does anyone know which systems take the lion share in a MLV? For example in a Falcon 9 (they are less secretive than others), how much goes to the engines, avionics, tanks, piping, structure, quality control, overhead, recovering hardware, tracking, etc.

If engines were the most expensive part, how much could the cost of a Merlin be reduced if they did a disposable version?

If every risk were unacceptable no ship would ever fly.

[RocketmanUS]

If not mistaken margin for cargo is 1.2 and crew is 1.4

High flight rates for RLV's or disposable would lower per launch costs. So the tanker and cargo launchers should be the same to have higher flight rates.

For Atlas the RD-180 was said to be around $10M each and the launch of the smaller version of Atlas V was around $138M

[go4mars]

No, you don't understand the concept.  It is not a cheaper payload, but cheaper launch vehicle.  Also, the cheaper  launch vehicles aren't blowing up left and right.  They are just simpler.

What important design choices make a launch vehicle simpler/cheaper? 

[savuporo]

What important design choices make a launch vehicle simpler/cheaper? 

Propellant type, pump type ( or pressure fed). These are the primary variables.

EDIT: Number of stages, obviously. The size of the rocket. Beyond a certain size, it gets very expensive due to ground infrastructure requirements.

And its all driven launch rate ultimately anyways.  No rocket flying once a year is going to be cheap.

[kkattula]

What important design choices make a launch vehicle simpler/cheaper? 

Higher structural margin, alowing cheaper materials and manufacturing techniques.

Lower payload fraction. Burns more fuel per mass payload, but fuel is usually very cheap.

[pathfinder_01]

What important design choices make a launch vehicle simpler/cheaper? 

Higher structural margin, alowing cheaper materials and manufacturing techniques.

Lower payload fraction. Burns more fuel per mass payload, but fuel is usually very cheap.

and less labor needed to launch.

[Ben the Space Brit]

I must admit that when I think 'tanker', I am thinking more 'robot in-flight refuelling vehicle'; something like a Cygnus or ATV with pumping equipment so it can refuel an upper stage in orbit.  That would require a degree of engineering and ground checks before launch.

However, it occurs to me that, if we go down the depot route, all you really need is a payload tank attached to an instrumentation, power and propulsion unit like the ATV or Cygnus SM.  The pumping equipment will be on the depot platform.  The IFR version would only be needed to support a vehicle that, for whatever reason (most probably launch inclination) would not be able to rendezvous with a depot platform.

[kevin-rf]

and less labor needed to launch.

Ding Ding Ding we have the Winner!

The standing army is the real cost driver. The only way to drive cost down is to use less people, or fly more often with the same number people. The only other option is to only hire interns

[edkyle99]

If we could determine a function for rate of success vs cost, and accepting cargo value is negligible against launcher value, should be possible to calculate the sweet spot where we can get the most mass in orbit for the least money.

The answer depends on how much propellent is needed and for how many years.  The propellant delivery rate.

A limiting factor would be the cost of the missions that would use the propellant!  The propellant could be free to its users but the missions themselves (lunar or Mars landings, say) could still cost too much to fly very often.  As a result, the total propellant delivery rate will likely be too low to support development of super-monster rockets, which would be the most cost effective way to launch heavy payloads but only if they could fly constantly and fairly frequently. 

For most purposes, then, the cheapest way to deliver propellant would be to skip the high cost of rocket development and simply use existing rockets - preferably lots of them.  Their per-unit costs should decline as their flight rate increases. 

 - Ed Kyle

[douglas100]

Ding Ding Ding we have the Winner!

The standing army is the real cost driver. The only way to drive cost down is to use less people, or fly more often with the same number people. The only other option is to only hire interns

Or buy your launch from a country with lower wage rates. What? It's already happened?

[mmeijeri]

MicroCosm has been the recent proponent of the concept, with their Scorpius launcher plans.

Space Systems/Loral's Aquarius Launch Vehicle concept is another example. I'm not sure Microcosm's launchers are intended to be unreliable, I think they're merely simple and well-suited to mass-production. They're similar to Sea Dragon in that respect, though obviously very small instead of enormous.

Sea Dragon: simple and huge
Scorpius: small, simple, mass-produced
Aquarius: small, simple, mass-produced, unreliable

[wolfpack]

Private space companies SpaceX and Blue Origin seem to think that recovering the LV is the best way to reduce cost. Not dropping millions of dollars worth of hardware into the ocean is the best way to save millions of dollars!

[mmeijeri]

That's certainly true for precious cargo, like aerospace hardware or humans. But for bulk materials it's not inconceivable that mass produced, unreliable and expendable launchers would turn out to be cheaper. Given a market for selling propellant in orbit the market would discover what the most economical approach was.

[edkyle99]

That's certainly true for precious cargo, like aerospace hardware or humans. But for bulk materials it's not inconceivable that mass produced, unreliable and expendable launchers would turn out to be cheaper. Given a market for selling propellant in orbit the market would discover what the most economical approach was.

Why plan to build "unreliable" launchers?  I'm not sure how making a launch vehicle less reliable would save money.  I'm not even sure how to save money by making a launch vehicle less reliable! 

 - Ed Kyle

[mmeijeri]

Why plan to build "unreliable" launchers?  I'm not sure how making a launch vehicle less reliable would save money.  I'm not even sure how to save money by making a launch vehicle less reliable! 

The Aquarius plan is to use a very simple single stage vehicle with very small margins, small enough that your vehicle can make it to orbit, but so small that one in three times it won't.

[sdsds]

I'm interested in the assertions about labor costs.

The trick for tankers will be to examine the flow of vehicles (reusable or expended) through the system and minimize the (mostly labor) costs to maintain that flow.  Personally I would target a system that launches every two weeks, where the staff for pad and control room operations consisted of only 10 workers and 4 managers working 40 hour work weeks.  Factor in on-going training, plus vacation and unplanned sick leave, and keeping a reliable launch tempo.  The result would need to be a fairly lean operation, with plentiful automation but also a fair amount of cross-training between positions.

I believe for the most part those workers need a bachelor's degree or equivalent and would expect around $90k in salary on average.  The overhead costs to employ them would likely double that, 14 * 90 * 2 ~= $2.5m per year.  If they performed 50 launches, that would be $50k per launch.

Is that estimate (a) reasonable; (b) off by a factor of 2 or more; (c) off by a factor of 10 or more?