Sorry, But These Numbers for a Launcher are Too Good to Believe

I smell snake oil, not LOX/RP-1

Elon Musk, is claiming some mind boggling figures for SpaceX's new heavy lifter:

Space Exploration Technologies, or SpaceX, plans to build a commercial heavy-lift rocket that will carry more than twice the payload of existing large rockets at one-third the cost. That would lower the price of delivering cargo to low-Earth orbit to the long-sought, and so far mythical, $1,000-per-pound range, the company's founder and chief designer announced today.

Speaking at the National Press Club in Washington, SpaceX CEO Elon Musk said the Falcon Heavy--made up of three Falcon 9 core stages powered by 27 upgraded Merlin engines and generating a combined 3.8 million pounds of thrust--will be ready for its initial test flight from Vandenberg Air Force Base, Calif., late next year or early 2013.

So he is claiming that his launcher will have a cost per pound to LEO ⅙ that of existing systems, and that he can have it ready for launch next year.

If you look at the prices quoted, they simply do not make sense internally either.

A quick perusal of the Wiki reveals that the Falcon 9, which is basically the Falcon Heavy core less the strap on boosters, puts 9,900 kilos into LEO for 44 to 49.5 million dollars, while the Heavy will put 53,000 kilos in LEO for 55 and 95 million dollars.

How do you manage to put 5½ times as much payload up for only 10-90% more?  My guess is that you don't.

The answer is you don't, not unless you come up with a way to bend the laws of physics or of manufacturing.

The Falcon Heavy will have 30 engines, 9 on each strap-on, 9 on the 1^st stage, and 1 on the 2^nd stage, so even if ¾ of the cost of a launch (the number is typically well under half) would be those 30 engines (it's probably less), the cost per engine would be on the order of 1¾ to 2½ million dollars each, which is very cheap.

I don't see how these cost targets can be met.