Tuesday, November 01, 2005

Why Merlin?

My anonymous commentor is getting at a very interesting point. Why does SpaceX do their own engines when clearly better engines are available from Russia? The Russian engines have far better Isp, burn the same practical fuel mix, are available for known and probably reasonable amounts of money, and are known to work now (which takes a lot of schedule uncertainty out of SpaceX's plans).

Using Russian engines was Kistler's plan. Kistler spent five times as much money as SpaceX has without building a complete vehicle. Maybe it was their recovery system. Maybe it was their basing plan.

I suspect that using a Russian engine puts a big Chinese wall in the middle of the company, for both intellectual property and ITAR reasons. Enough of the vehicle design is tied to the engine design that what you end up with is a company that, to a noticeable extent, resells Russian launch services at the whim of international relations.

How big a deal can this be? The Atlas-V uses Russian engines, and is intended to be used by the military for sensitive launches. If Lockheed can do it, why not SpaceX?

According to Astronautix, Atlas V has launched 4 communication satellites, only one of which was a U.S. satellite, which was commercial. By comparison, Delta IV has launched 2 U.S. military commsats and a european commsat. My sense is that the U.S. military is averse to relying on a launcher using unsubstitutable components from a major overseas competitor, and funds the Atlas solely as a backup to the Delta.

I think SpaceX knows it will be dependent on launching U.S. military payloads, and knows it can't do that with Russian engines.

If SpaceX is sucessful, I expect an EELV, probably Atlas, to get cancelled by 2010. Delta will become the backup launcher, subsidized by the military and flying in very low numbers.

And seven years from now, I think the boost competition will not be United Space Alliance versus Ariane versus SpaceX. I think it will be SpaceX versus at least one Russian company (perhaps marketed by a western company e.g. SeaLaunch) versus the Chinese. SpaceX will get the U.S. military business by default. To win the international competition, SpaceX may need to figure out reuse.


  1. I think SpaceX will do fine without touching NRO or USAF payloads. Bigelow will need 15 Falcon 9's and NASA will need many more to finish the Station (28) and obviate the need for the HLLV (???).

    -Tony Rusi

  2. From a purely technical point of view, it does indeed not make any sense to build your own engine. No disrespect to the spacex engineers, but the NK33/43 engines are better than anything spacex engineers can afford to develop anytime soon.

    But this is not a purely technical problem. SpaceX have to produce the most critical parts of their launcher in house to be safe from meddling by BoLockMart.

    Strongarming by the two big competitors is so severe that spacex does not have a reliable source for AlLi, so the payload for the Falcon I is 100kg less than it could be. If they did not have full control over their engines, they would probably be out of business by now.

    So from an overall strategy point of view, doing their own engine was a good decision.

  3. I have another thought about why SpaceX may be developing their own engines in house. From the beginning, Mr. Musk has said that he wants to change the launch industry by designing his rockets to be more reliable and to fly more often than anything out there. I think he's accepted as an axiom that the more complex the design, the more modes of failure that exist, not to mention that it probably makes the engines more expensive to manufacture and maintain. So, the crux of his of his strategy thus far seems to have been to design the simplest possible vehicle that can get the job done while still being mindful of cost and reliability.

    Now, while there may be some very good engines out there, with very good track records, it is possible that they are slightly more complicated or expensive than SpaceX needs to get the job done. In addition, it may be that by developing their engines in house, they also have control over the production rate and quality control. It may very well be that there were design decisions made for the Merlin engines that will allow their production rate to be scaled up to match their flight rate.

    By making these decision in house, SpaceX is insulating themselves from the engine manufacturing market. In this way, they can have a steady supply of engines without having to rely on any one else to supply them, or as anonymous mentions above, without worrying about any unfair practices from existing competitors.