The commercial space flight business suffered a one-two punch on October 28. An unmanned rocket carrying supplies for the International Space Station and launched by Orbital Sciences Inc failed a few seconds after launch falling back to the launch pad and exploding to make a spectacular night-time video that must have been shown on every TV outlet in the US. It was the company’s third commercial launch of a contract to supply the Space Station, whose residents will now have to wait a while longer for the next garbage pickup. (A side benefit of the long-distance unmanned deliveries is that the Space Station folks can cram the vehicle with their trash and let it burn up in the atmosphere.)
And then on October 31, Virgin Galactic’s SpaceShipTwo, manned by two experienced test pilots, broke up high above the Mojave Desert in California, killing pilot Michael Alsbury, 39, and injuring the other, Peter Siebold. The crash scattered debris over a five-mile-long area and initiated an investigation by both Virgin Galactic and the US National Transportation Safety Board which could take as long as a year.
Any time anyone is injured or killed in a space-related accident, engineers are obliged to get to the bottom of the technical whys and hows of the mishap. But beyond the specific technical causes of these particular accidents, tragic as they were, is the question of how reliable commercial manned space flight is going to be. And a little history can throw some light on that question.
A man named Ed Kyle maintains an extensive statistical study of space-flight launches at a website called Space Launch Report. He compiles both unmanned and manned flights, although in the nature of the business, the vast majority of launches are unmanned. Bearing that in mind, we can look at a convenient summary table he provides of success rates of launches by decade, going all the way back from the infancy of space flight in the 1950s to the 2010s.
America’s first attempt to launch a satellite into orbit, the Vanguard launch on December 8, 1957, was a highly publicized failure, exploding after reaching the breathtaking altitude of four feet (1.2 meters). And overall, only about half the launch attempts by all parties in the 1950s were successful. But aerospace engineers began climbing that long haul called the learning curve, and by the 1970s the average success rate was around 95 percent, where it has hovered ever since. In the last two complete years, for example (2012 and 2013), Kyle logged 159 launch attempts and nine failures among them, for a failure rate (for the pessimists among us) of 5.6 percent. So even today, 40 years after the space-rocket business reached maturity, there is about one chance in 20 that your satellite will not end up in space, but in a watery or earthy grave.
Despite all the fuss about NASA turning space flight over to commercial interests, satellite launches have been commercial transactions for decades. And it appears that a failure rate of 5 percent is an acceptable level to support a generally prospering space industry. The companies and their insurers can handle that level of failure and still accomplish what they want to do, most of the time.
But launching cans of beans for a space station, and launching people who have paid a quarter of a million dollars for the ride (as prospective passengers in the Virgin Galactic rocket have coughed up in advance), are two different propositions. Commercial airlines would not have many customers if it were well known that one out of every twenty flights was going to crash. It took the business of aviation twenty years or so to be safe enough to offer commercial passenger service, but by 1930 or so the risks of commercial scheduled flights to the individual passenger were largely imaginary, and today you take more of a risk of dying on your drive to the airport than you take in the air.
It may be harder for the space-flight engineers to drive their failure rates down to the level at which people could buy space-flight life insurance for a few dollars, like you used to be able to do for commercial aviation flights at airports. Rocket hardware operates at the outer limits of materials science. The engines run so hot that liquid-fuelled nozzles have to be cooled continuously to keep them from melting, and the fluid dynamics of the combustion of rocket fuel is still so complex that an exhaustive, essentially complete mathematical model of a rocket in flight, including vibration modes and so on, is quite possibly still beyond our abilities. So rocket designs are a combination of science-based modelling and engineering intuition, added to a large measure of experience of what has worked in the past.
I think it is significant that the Virgin Galactic flight was using a different type of fuel than they had used in previous flights. Such a major change, even if tried out on the ground with similar hardware, can lead to unpredictable results, and may turn out to have contributed to the disastrous crash of SpaceShipTwo. Rocket engineers, at least the successful ones, tend to be highly conservative in their designs. Anyone who has seen both an old V-2 rocket engine in a museum and the massive Apollo engines used to launch men to the moon can see that Wernher von Braun found something that worked at Peenemunde, Germany in the 1930s, and stuck with it all the way through the 1960s.
Such conservatism is increasingly rare among engineers in general today, influenced by innovations in hardware and software which happen so fast that you can squeeze an entire product life cycle, from introduction to obsolescence, into six months. But the adage “if it ain’t broke, don’t fix it” applies in spades to space travel. And as we find out in the coming months what caused SpaceShipTwo’s failure, we may find that experimenting with a different fuel was a bad idea.
Unless we colonize the Moon or Mars to a great extent, space travel will always be an exotic, low-volume business, like tours to the Antarctic are today. And it is by no means clear to me that even the super-rich will be willing to take the kind of risks that simple statistics tell us space travel entails—at least, not for quite a while yet.
Karl D. Stephan is a professor of electrical engineering at Texas State University in San Marcos, Texas. This article has been republished, with permission, from his blog, Engineering Ethics, which is a MercatorNet partner site.