Musketeer Mike usually covers the Elon Musk beat, but he is on vacation so the rest of us will fill in. And this one is appropriate for me because I am an architect and am used to trying to avoid catastrophic failures, that's part of our job. With the SpaceX flight, it appears that a little strut holding up a helium tank broke. It's a two foot long piece of steel that was designed to handle 10,000 pounds of force but snapped at 2,000. According to Verge, "Musk said SpaceX still doesn’t know why the steel rod snapped, but it’s possible that its material was faulty. Musk didn’t name the strut's supplier, but did say it may be that one strut of thousands wasn’t up to code."
In a later update this afternoon, Musk says "SpaceX will switch to different struts with even better safety records."
"“Obviously, what we’re going to do in the future is not use these particular struts, and we’re going to move to individually testing each strut independent of any material certifications,” Musk said. “There will be some cost increase to the rocket, and not we believe to a significant amount."
Now back to the architecture reference. Architects and engineers are trained to minimize risk. We have others check our work, then building examiners check that before permits are issued, then building inspectors looking at it after that. Then we and our engineers have to sign off on it as well and carry a ton of insurance. Going into space, it is the same thing magnified a hundred times, yet so many disasters have happened because it wasn't done. Apollo 13 because of a
faulty soldering job short circuit caused by a sloppy repair. Challenger because of a crummy design that people knew was problematic but went with anyway.
When a balcony collapsed in Berkeley a few weeks ago, killing six students and injuring seven others, it was perfectly obvious to anybody who knew anything about construction that this was a crummy detail that eventually was going to fail; the question was only whether inspectors would catch it before it actually did. Yet thousands of designers use that detail because it is cheap and easy.
The lesson from both these events is the same. You have to design it right in the first place, you have to test, you have to inspect, you have to maintain, you have to pay enough to have it done right. If only as much attention was paid to a failed balcony as was to a failed rocket.