Rocket Science

(In a conversation elsewhere, I responded to complaints about the “NASA rocket” exploding — and to suggestions that this stuff was not new technology and should be easy by now.)

It was not, exactly, a NASA rocket. And it was indeed new technology. This was a launch contracted from NASA by Orbital Sciences, who have been having some trouble with their systems. Enough trouble to cause them to change the name of the rocket (from Taurus to Antares, kind of “anti-Taurus”) to avoid bad vibes/publicity/luck. SpaceX is doing rather better, so far.

Orbital’s engines are purchased from Russia (more below on this), but everything north of that is of very recent vintage. And all of this rides on the edge of the performance envelope; every ounce added reduces payload capacity by $1,000 or so.

These are extremely complex systems, made more likely to fail by the demands of NASA bureaucracy supposedly to prevent failure. A little patience is needed, still. Or a review of the many “failure reels” on YouTube of NASA’s own efforts. Failures continue even this year.

Why Russian engines?

Suspicion has fallen pretty quickly on the Antares’ A-26 engines. “Oh, they’re Russian. That explains it.”

Well, to be fair, the Russian engines have been pretty reliable in recent decades. For various reasons, the US lost much of the multi-axis machining skill needed to make new engines, from mothballing the tools to retiring the machinists who knew how. (Disclaimer: 40 years ago, I was making Space Shuttle parts.)

Several companies have recently been designing and building new engines, but there has been a lot of “start from scratch” in this effort. Many of the “NewSpace” companies of a decade ago (Kistler, for example) were using retread Soviet ballistic missile engines. The Rotary Rocket company had a really clever new design … but the company failed, sadly. How many spaceships do you know of with helicopter blades? I was privileged to be involved in that effort, if in a minor way: The streaming of the rollout was facilitated by my company, as well as the sending of press-kits and other materials about this unique approach to space. This video recaps the company’s work:

But these engines purchased from Russia for the Antares are not the same engines that the US has been buying for NASA use. More on that in a moment.

An expensive failure

The mission was insured for about $200 million. However, it was not carrying an exotic, expensive satellite payload; instead, it had nearly a ton of various small instruments and satellites, including a bunch of student experiments.

Insurance companies are used to multi-hundred-million-dollar launches, but this one does not seem to rate that. And launches must become cheap if we are to expand from this planet. It was expensive in other ways, too: Orbital’s stock price also fell abruptly.

Wired has a discussion of the payload, and they attempt to play up the loss as a big hit to science:

Last night, an unmanned Antares rocket carrying more than 5,000 pounds of cargo to the International Space Station exploded in a huge fireball seconds after liftoff. No one was hurt, but in addition to damage to the launch pad, NASA lost tons of supplies, including equipment and food for astronauts.

Science also took a big hit in the explosion.

Almost a third of the payload (by weight) consisted of science experiments that ranged from a student project studying how pea shoots would grow in zero gravity to a high-tech camera that would have been the first to monitor meteors from space.

The ISS had enough supplies on board to last the crew until March of 2015. Also, a Russian Progress resupply mission launched later this week.

SpaceX has a resupply mission scheduled for December, and it is likely to have its cargo manifest altered to address some items lost this week.

Not a joke

My friend commented that Elon Musk of SpaceX had once described Orbital Sciences’ design as a joke. That article is here, and it refers to a comment he made during a 2012 interview:

One of our competitors, Orbital Sciences, has a contract to resupply the International Space Station, and their rocket honestly sounds like the punch line to a joke. It uses Russian rocket engines that were made in the ’60s. I don’t mean their design is from the ’60s—I mean they start with engines that were literally made in the ’60s and, like, packed away in Siberia somewhere.

One should read this keeping in mind Musk’s own words, noted when his own rocket design exploded two months ago:

The exploded rocket was a Falcon 9R, the model that successfully delivered the Dragon spacecraft to the International Space Station in 2012.

The Falcon 9R rocket is a successor to the Falcon 9, which was nicknamed Grasshopper because it can land upright on retractable legs.

SpaceX’s founder Elon Musk had apparently expected malfunctions in developing the new technology.

“I do think there probably will be some craters along the way; we’ll be very lucky if there are no craters,” SpaceX founder Elon Musk told the Royal Aeronautical Society in 2012.

A Dangerous Dependency

Incidentally, the Antares first stage (the portion that exploded) is not using Soviet ballistic missile engines. It uses the old Soviet N-1 Moon Rocket engines manufactured during the Apollo era space race. And the entire first stage is built in the Ukraine, then shipped here.

There are US government launch vehicles using Soviet/Russian ballistic missile engines, and the current antagonism between the countries risks cutting off our supply of those engines. This is a dangerous dependency, in my opinion. Here’s a Senate hearing on the topic, where we learn that (unsurprisingly) we are very poorly prepared for such events:

Where are the rocket scientists?

Orbital Sciences, Blue Origin, SpaceX, XCOR and others are strongly attractive to the crew of rocket scientists these days. These NewSpace companies are almost the only game in town, as NASA is no longer prominent in this arena, and the giant firms have an interesting policy that makes it difficult for the seasoned hands to get placed there.

In other words, they’ve got rocket scientists, and rocket engineers, and people who qualify as both — they can both burn up the chalk and bend the metal.

I’ve met more than a few of these folks, and many are impressive indeed. But the job is an enormous one, and this is especially true on work that is done under contract with hidebound NASA. An old saying for government contractors: You’re cleared to launch when the weight of the paperwork equals the weight of the vehicle.


As Elon Musk said, “Rockets are hard.” Even NASA, with a huge budget to get it right, still had a one-in-fifty failure rate on the Shuttle. And that was a man-rated system.

As Richard Feynman discovered in the aftermath of Challenger, that failure rate seems to have been great good luck; the system was really quite fragile. The cracks in the turbine blades, the leakage of O-rings, and other issues were always seemingly on the edge of disaster.

I participated in a minor way in the engineering analysis after the Columbia re-entry disaster, and the same thing was evident there. It “should” have failed much more than it did; we were lucky. In that case, the focus was on all the damage being done by ice at liftoff, by debris on-orbit, and by burn-through during re-entry. In one case the pivot rod for the elevon (a control surface on the wing) had burned almost completely though; a bit more, and the orbiter would have been uncontrollable during re-entry and landing. And there were thousands of holes punched in the heat shielding, some near-critical, on every flight.

Even carrying the orbiter back to Kennedy on the back of a 747 was dangerous: Rain drops at 300 MPH were enough to chew up the ship’s heat-protecting tiles, causing the need for a massive replacement. That transport craft was always preceded by fighters ranging ahead and making sure there was no rain in the path, and picking a route around the clouds.

This sensitivity and fragility were caused by the decision by NASA, late in the design process, to eliminate the titanium shield over the skin that would have protected it. In so doing, they reduced the spacecraft’s weight by about 10,000 pounds (an increase in payload by nearly 20% on an equatorial orbit), but added greater risk to each mission. Such trade-offs are common, and the results are sometimes unforgiving.

===|==============/ Keith DeHavelle