The Bad News.

Suggested as a possibility several months ago, NASA decided last Saturday that the two astronauts sent up to the ISS on June 6 for the first crewed test flight of Boeing’s Starliner spacecraft, Suni Williams and Butch Wilmore, will return to earth in February 2025 on a SpaceX Crew Dragon spacecraft. The Starliner will now undock and return to Earth by remote control on September 6.

So not the success NASA and Boeing wanted.

It also means some buggering around with the SpaceX Crew-9 mission; starting with a delay of its launch by a month to late September to allow time to get Starliner off the docking port; getting two spacesuits for Williams and Wilmore since their Starliner suits won’t work with Crew Dragon; and reducing the launch crew by two to make room for the Starliner pair on the return trip. I’m a little surprised by the latter since Crew Dragon was designed to hold seven people, has been reduced to six and typically flies four but can fly the original seven. Why not just send up the full, scheduled crew for the ISS?

Admittedly that would push the ISS crew from seven to nine and extending that for six months would complicate the consumables planning for food, oxygen and water. But I’d have thought the ISS is big enough to have that sort of spare capacity, and they could schedule up extra supply ships?

Oh well, the news is not entirely bad since it means that at least NASA won’t have two dead astronauts on its hands and hopefully not even a destroyed spacecraft. Starliner should land in the New Mexico desert where it can be studied and the problems fixed for the next test flight.

As an aside it’s interesting that Boeing abandoned the American practice of water landings for such spacecraft and went the route the Russians have always used. It was a sensible choice for the Soviets since they had a vast land area with limited, suitable water areas around it, a small navy, and a paranoia that would never have allowed them to land in the Pacific or Atlantic oceans where the US Navy reigned supreme. Boeing had the same, easier options that NASA had in the 1960’s, and with better re-entry and landing precision they could land closer to shore. I assume the driver of the design decision was the greatly reduced logistics support. Having said that SpaceX lands so close to shore that rubberneckers in ordinary pleasure boats have cruised out from Florida to watch them land, and their logistics support is tiny compared to what was needed from the US Navy for NASA’s Mercury, Gemini and Apollo missions.

One more piece of bad news for NASA comes on a more mundane front but still symbolic of the management problems that have plagued the outfit for decades:

In 2019, NASA awarded a $383 million contract to Bechtel to design and build a second mobile launcher that will be used to transport the SLS rocket to its launchpad for the upcoming Artemis 4 mission, scheduled for launch in September 2028. At the time, Bechtel was supposed to deliver the launcher by March 2023.
…
Despite the initial cost projections, the OIG report (NASA’s Office of Inspector General) estimates that the mobile launcher could end up costing $2.7 billion and that it would not be ready to support the SLS launch until September 2029.

Sigh! I had no idea they were building such a thing. Aside from the cost overruns, by the time it’s operational the whole Artemis program may well be running entirely on SpaceX’s Starships. The SLS continues to live up to the label it’s critics applied to it years ago: the Senate Launch System, designed mostly to funnel money into certain states where NASA and NASA contractor factories remain from their Apollo days.

The Good News.

In making this decision NASA has clearly (and finally) learned from its mistakes in the past, specifically the two Space Shuttle losses of Challenger in 1986 and Columbia in 2003. In both cases engineers issued warnings but management overrode them.

With Challenger the engineering uncertainty was whether the below-zero cold of the January Florida launch would damage the solid rocket boosters, causing them to fail. They had evidence from previous flights in cold weather where the seals between the segments had almost failed, but they could not conclusively say that would happen this time. That was enough for the management – still keen at that stage to achieve NASA’s original target of a Shuttle launch every month (or even faster) – to override them. The result was a destroyed Shuttle, seven dead astronauts, a massive blow to their reputation in front of the American public and two and a half years in limbo as an investigation was done and modifications made to the boosters and launch procedures.

In the case of Columbia the engineers faced another uncertainty: this time the question being whether the leading edge of the port wing had been damaged by a piece of foam insulation breaking off the big fuel tank. They’d seen it on the video (not 1080p quality, let alone 4K) and clearly it had hit the wing. But where, and how badly? They wanted to get a spy satellite to take pictures of Columbia but the management nixed that idea and continued with the mission. The result was another destroyed Shuttle, seven more dead astronauts and another two and half year suspension of the program.

This time the consequences were more severe as the investigation concluded that the Shuttle was a flawed system that could never be made as safe as necessary, with mere mitigation of the potential fatal risks. In fact it could not even have been rescued in space. So although flights had to be resumed in order to finish building the International Space Station (ISS) the Shuttle program was to be decommissioned as soon as that was done. Those recommendations were followed and the program ended in 2011, with the final flight being the last construction component of the ISS.

I’m sorry to say that the management failure in the case of Columbia didn’t surprise me, as I wrote about my visit to Mission Control in Houston in late 1986, almost a year after the Challenger disaster:

I also was not impressed by our tour guides response to questions about the disaster: “Look, we take risks and this time it did not come off“.

Since the Rogers Commission report on the disaster had already come out a few months earlier and I’d read it, that answer pissed me off big-time.  The report had pointed out that there was a hell of lot more involved than just the usual risks of spaceflight, including some very deep, structural problems with NASA.

I came away with the distinct impression that if these guys were symptomatic of the rest of the outfit then they probably hadn’t learned the real lessons.

This time it seems that they have. In announcing that the launch crew would not return on Starliner, none other than NASA commissioner, Bill Nelson, had this to say of the reasoning in response to a direct question as to whether the Shuttle accidents had played a roll:

It has affected the decision today by this collective group and all of those that participated in the Flight Test Readiness Review this morning. It is trying to turn around the culture that first led to the loss of Challenger and then led to the loss of Columbia, where obvious mistakes were not being brought forth.

I do have to wonder how much Nelson’s personal experience may have counted in this, because he flew on the Columbia in a January 1986 mission that immediately preceded that of the loss of Challenger. It could have been him who died on a Shuttle mission.

But in any case good on NASA. Yes, it’s a failure and an embarrassment for both them and Boeing (mainly the latter), but that’s better than two dead astronauts.