I’m listening in live now, but Leonard David already has a story up.
[Update a few minutes later]
Doug Messier has the executive summary. Full report won’t be available for a couple weeks.
[Update a while later]
Temporary adjournment to wordsmith final findings. Sumwalt wants to make clear that while it was pilot error (Finding 1), it was an institutional failure at Scaled that made it possible for such an error to be catastrophic. There also seems to be some (IMO, undue) criticism of FAA-AST, and talk of “political pressure” to prematurely issue licenses/waivers. People (including NTSB) need to understand that AST currently has no statutory authority to regulate safety of the spaceflight participants, including crew, and they are chronically under-resourced to carry out the responsibilities that they are authorized to do. In fact, George should be careful what he wishes for, because if the learning period expires this fall, he still won’t have the budget he needs to expand his authority.
One thing that’s not clear yet (to me): If part of the problem was inability to read instruments due to vibration under thrust. If there was a digital readout, that (and other critical information) should be replaced with an actual Mach meter.
[Late-afternoon update]
Here‘s Jeff Foust’s report.
Good morning, Rand. I’ve been waiting to hear your thoughts on the hearing.
When you say an actual Mach meter, you mean an analog dial and needle sort of dealy? Is it much more difficult to look at a display with the digits 1.3 as opposed to a needle pointing at 1.3? This investigation has piqued my interest in human factors engineering, but I have much to learn.
Also, it was suggested multiple times that inspectors be assigned to vehicles, instead of launches. Apparently there’s precedent for that in aviation. What are your thoughts on that?
It could be an electronic display, but yes, it should be a dial and needle or dial and bar. Much easier to distinguish in a vibrating environment than reading jiggly digits.
Thanks for weighing in. And assigning inspectors to particular vehicles? Is that a good idea? I believe Captain David Lawrence said during the meeting that this is standard procedure for commercial aircraft.
Rand, quick follow up on this tweet from your book account:
Other surprise that came out of this is that Scaled didn’t consult with FAA prior to design of SS2. Recommendation to not do that.
Why would they do this? What does the FAA know about designing spacecraft? Mr. Zoeller, in reading the proposed recommendations, suggested the FAA could offer human factors advice. Is that the sort of consulting information you had in mind?
What does the FAA know about designing spacecraft?
FAA-AST knows a lot about designing spacecraft. It wouldn’t be possible for them to judge the probability of damage to third parties without that. Yes, FAA could offer advice on human factors, but that’s not currently as much in their skill set, and they’re not currently resourced to carry out the NTSB recommendations.
Please take this as the very ignorant question it is: What I have read (very briefly) is that some of the human factors stuff they are being criticized for is not space related and is known in “other industries”, presumably aircraft. The kinds of things I would expect from (not to slander anyone) Armadillo or someone, people that do not necessarily have a background in aircraft best-practices. Scaled Composites should have that knowledge. Just the habit of ordering readouts as dials rather than digital because “that’s how we’ve always done it” (something that makes a lot of sense now that it’s pointed out but I would never have thought of) seems like something they would know to do. Was there any mention of why it wasn’t done? Or if they did it some other way intentionally and it didn’t work?
I’m not sure a dial actually solves the issue because it’s read relative to hash marks that are labeled 1.0 1.1 1.2, etc. You have to be familiar with the Mach meter to automatically know the needle angle/Mach number relation. If there are key critical Mach numbers then I would suggest a couple of large, bright idiot lights with different colors.
Having programmed lots of machinery and robotics, I’m also a fan of interlocks that prevent operator actions that will automatically result in death or bodily harm, because everybody occasionally has a brain fart.
I was assuming there would be yellow and red lines.
My skydiving altimeter has a dial, yellow and red bars. Never read it wrong, or never read it substantially wrong.
I’m also a fan of interlocks that prevent operator actions that will automatically result in death or bodily harm, because everybody occasionally has a brain fart.
+1