Tuesday, January 21, 2020

Boeing's Starliner Test Flight Went Worse Than We Thought

Boeing's Starliner test mission to the ISS a month ago aborted its main objective when a timer error caused it to miss its critical orbital insertion burn.

Today, Ars Technica reports that the mission also had failures in its thruster performance. The mission failure reported at the time that because of the timer error, the thrusters on the service module fired much more than the mission profiles call for.  They're now saying that, “Many of the elements of the propulsion system were overstressed.”
The NASA [initial information] release did not mention thruster performance, but an agency source told Ars that engineers are looking closely at the performance of the Starliner propulsion system. In addition to four large launch abort engines, the service module has 28 reaction control system thrusters, each with 85 pounds of thrust and 20 more-powerful orbital maneuvering thrusters, each with 1,500 pounds of thrust.  [Note: text in brackets added - SiG.]

During the post-flight news conference Jim Chilton, Boeing's senior vice president of the Space and Launch division, said the service module thrusters were stressed due to their unconventional use in raising Starliner's orbit instead of performing one big burn. As a result, the company had to shut down one manifold, which effectively branches into several lines carrying propellant to four thrusters. "We even shut down one manifold as we saw pressure go low 'cause it had been used a lot," he said.

The NASA source said eight or more thrusters on the service module failed at one point and that one thruster never fired at all.
 Ars asked for a comment from Boeing, and they were given the following statement:
"After the anomaly, many of the elements of the propulsion system were overstressed, with some thrusters exceeding the planned number of burns for a service module mission. We took a few cautionary measures to make sure the propulsion system stayed healthy for the remainder of the mission, including re-pressurizing the manifold, recovering that manifold’s thrusters. Over the course of the mission we turned off 13 thrusters and turned all but one back on after verifying their health."
In my opinion, after Sunday's successful SpaceX test, this puts Boeing a bit behind the eight ball.  After all, the emphasis up 'till now has been (my interpretation) trying to determine if they could claim the mission met all requirements.  Since the Starliner capsule never completed its primary mission of docking at the ISS, it seemed to be a stretch and everything else had to be superb.  With these apparent thruster issues, it seems to me Boeing would have to show the specifications for the thruster system and demonstrate, line by line, how they proved it worked per the specifications and here I'm assuming the design went through Critical Design Reviews where both Boeing and NASA signed off on the requirements, agreeing to them.

Two weeks ago, NASA and Boeing committed to a two month schedule to complete the analysis of the timer anomaly, while concurrently determining if another test flight is required.  My bet is that they'll have to redo this test flight essentially in its entirety.


Starliner on the pad on 12-19-19.  Trevor Mahlmann photo.

During Sunday's post-testflight press conference, NASA Administrator Jim Bridenstine repeatedly emphasized that he wants more than one ride to the ISS.  He wants both Boeing and SpaceX to succeed and he wants a vibrant private sector in space.  Everything he said sounded reasonable to me.  Both Bridenstine and Elon Musk commented that they could have the Crew Dragon ready to fly to the ISS by mid-March, but that with other considerations on the table they collectively said “second quarter.”  I'll be keeping my ears and eyes open for any changes.



13 comments:

  1. Better these failure happen now in testing rather than later in commercial use.

    ReplyDelete
  2. I like the way SpaceX works. They are their own worse critics, and they don't try to cover up flaws or performance issues.

    Boeing, on the other hand... You can't tell me they knew exactly the performance issues they were having, as it happened.

    Any more, since the late 80's, these rockets are so wired for telemetry that it's almost impossible to not know within a few days what went wrong.

    If SpaceX can approximate what's wrong almost right away, and give you a definite answer almost right away, and identify a fix within a week and have that fix implemented and tested within a month, then why can't Big Aerospace do the same?

    ReplyDelete
    Replies
    1. Boeing is afflicted with Big Defense Contractor's disease. I think it comes from living on "Cost Plus" contracts for so long. I retired from a smaller company (we'd be a subcontractor to Boeing) and used to joke that if the building caught fire, it would take us 6 months to evacuate the place.

      These companies get infected by needing pounds of paperwork for everything they do, to the point that nothing can be done without 500 approvals.

      Delete
    2. My main career was at Boeing. You need to find a different joke, as we had building evacuation drills once every month or two, and it usually took less than ten minutes to get everybody out and verified.

      What did usually take six months was finding a manager to sign off some particular critical piece of paper so you could procure something. Paper clips or a building-size quantum stellar vacuum vortex magnetron, it didn't matter.

      I didn't appreciate the half-mile walk to the nearest bathroom, either.

      Delete
    3. When I worked at Sea Launch we joked that the rocket wouldn't lift off until the pile of required documents was as tall as the rocket.

      And never ONCE did I see somebody pencil-whip a document or test to pass.

      Delete
  3. SpaceX seems to be out-pacing Boeing.

    ReplyDelete
  4. NASA Administrator Jim Bridenstine repeatedly emphasized that he wants more than one ride to the ISS. He wants both Boeing and SpaceX to succeed and he wants a vibrant private sector in space.

    Lies. NASA is gun control for ICBM's, pork for weapons maker cronies, and an iron curtain keeping the tax cattle from escaping to the new frontier.

    ReplyDelete
  5. Of course they had failures in their thruster performance! They don't TEST anything. The main contractors don't spend enough time with parts on a test-pad: They're busy chasing this fantasy that their CAD systems and modeling are so sophisticated they can have their vehicle spring forth complete from their heads like Athena. They spend their time sending each other powerpoint slides and gant charts, and there is a strict separation between the information starved engineers, and the technicians at a bench that are supposed to build this thing.

    There's far more rant where this came from.

    MadRocketSci

    ReplyDelete
  6. PS: My rant is mostly inspired by my time elsewhere, not a Boeing. A similar project, run in a similar way, doomed for what I assume from my experience are similar reasons.

    If what I read is correct, as near as the 90s, Boeing would obsessively test parts and sub-components. They would make a part 15 times, adjusting the way they would perform a weld, then x-ray them for flaws.

    During the Apollo program, they built several versions of the command module. Everything was tested. They iterated on the design extensively. There was a very incremental program of ground-tests, launches, in-space-tests, manned and unmanned building up to the eventual moon landing. Nothing like our space program today where SLS-1 is supposed to be "the test", and SLS-2 has astronauts on top.

    Testing is "expensive". Spinning your wheels for years generating powerpoint charts and models (many of which need inputs that no one can give you) is "efficient".

    MadRocketSci

    ReplyDelete
    Replies
    1. I don't have a problem with any of what you say. I think we all know that there are things that simulation is good at and things that really need physical testing. Knowing which results to believe and which to question is what separates the guy with 20 years experience from the guy with 1 year of experience 20 times. Models have to be validated by physical experiment. You don't just sit new grads, or 2 year guys, or really anybody down at a computer and fly their results

      I think the first space vehicle that's first flight was manned was the Space Shuttle. There's a video out there with the two man Crew (John Young and Bob Crippen - who joked they thought by the time they flew they'd be 'Old and Crippled') talking about the nerve-wracking tension that brought. I remember Crippen saying, "when they lit those solid rockets, we knew were going somewhere."

      Delete
  7. Perhaps I over-simplify, but it seems to me that Boeing has went from a company ran by engineers to a company run by bean counters. I have never found a company run by accountants to be one I cared to deal with.

    ReplyDelete
    Replies
    1. There's some hold out "Old Skool" divisions left, but they're getting fewer and fewer as the older guys retire.

      Delete
    2. Remember that the true customers of Boeing (and any publicly traded company) are the major stockholders. Those that purchase there products are secondary.

      Delete