Actually, it IS Rocket Science

...

Chincoteague, we have a problem

So last week we saw two spectacular, if unrelated, failures in the commercial spaceflight realm. What's the deal? One was a routine cargo flight to the ISS in low earth orbit, and the other was an atmospheric test of a commercial sub-orbital spacecraft being developed to serve the space tourism market. We've been doing this kind of thing for well over forty years now. Don't we have it mostly figured out at this point?

The fact is that both of these accidents represent the kinds of catastrophic failures were going to continue to see in spaceflight forever, no matter how mature the technologies and routine the processes become. In the case of the Orbital Sciences Antares launch vehicle, it was very likely a failure of the liquid fueled rocket engines used in the first stage, and in the case of SpaceShip 2 it appears to be either a pilot or systems error in the deployment of an experimental set of aerodynamic surfaces intended to facilitate deceleration and control upon re-entry.

Much has been made of the fact that the Antares first stage used the Aerojet AJ-26 bipropellant ligquid fueled rocket engine, which is a refurbished Soviet era NK-33 rocket originally designed and built for the Soviet manned lunar program. But that's unfair. These engines (with one significant exception we'll get to in a moment) are mature, thoroughly tested designs that have been successfully employed for decades. Rocket engines are very hard to get right, and both liquid and solid fueled engines have their own complexities. But large liquid fueled rockets are notorious - they are essentially huge, highly sensitive bombs that are - if all goes well - detonated in a controlled fashion, generating lift based on Newton's third law.

The NK-33 is a staged combustion design, using Liquid Oxygen to oxidize a high grade kerosene called RP-1. In the preburner stage above the turbopumps, it uses an unusually oxygen rich fuel to drive the turbopumps. Most rocket designers have shied away from that approach, as the oxygen rich fuel is highly destructive to metal, and can result in burn-through. We don't know if this is what happened with CRS Orb-3, as the Soviet metallurgists have historically successfully prevented it, but it is worth consideration. Also worth noting is the use of the Castor 30XL second stage for the first time on an Antares launch vehicle. While there's no reason to initially suspect the second stage in the loss of thrust event that resulted in the destruction of the spacecraft, it certainly contributed a great deal of powerful fuel to the explosion.

For the loss of SpaceShip 2, the cause may be even more prosaic. The spacecraft uses a deployable set of aerodynamic surfaces as part of its re-entry process. For some reason, the pilots released the lock and the system, called 'Feathering' deployed during powered climb rather than later, on unpowered descent. This may have rendered the craft uncontrollable and led to the crash.

But the key takeaway from last weeks catastrophes is that chemical rockets are very hard to design, build and operate, and that space travel will never be 'safe'. The very idea that you could consider yourself safe sitting on top of a giant bomb intended to shoot you out of earth's atmosphere, and then somehow returning that hyper-velocity craft to the surface in one piece is kind of delusional. But that's ok - people who seek safety and people who seek adventure are never the same people, and before you get where you're going you have to expect some pretty tough setbacks. I hate it when we lose good people, but I'm glad there are people willing to work on these problems despite the costs.
...