This is not a NASA Website. You might learn something. It's YOUR space agency. Get involved. Take it back. Make it work - for YOU.
Commercialization

Boeing Starliner Pad Abort Test Was Technically A Success – But …

By Keith Cowing
NASA Watch
November 4, 2019
Filed under , ,
Boeing Starliner Pad Abort Test Was Technically A Success – But …

Keith’s note: The pad abort test of Boeing Starliner was technically a success today. The system quickly removed the capsule from the danger zone and landed exactly as planned but one of the three main parachutes did not deploy. The NASA and Boeing TV announcers repeatedly commented that 2 deployed parachutes are within the safety requirements of the system, that this is all about redundancy, and that a safe landing could have happened with one parachute. But one of the three main parachutes failed to deploy. Given previous parachute problems, it is possible that additional testing will be required before Starliner who can be launched. Boeing was originally not planning to do a live broadcast of this test until NASA Administrator Bridenstine told them that they were going to do it.
Keith’s update: NASA’s post- test press release says “Two of three Starliner’s main parachutes deployed just under half a minute into the test, and the service module separated from the crew module a few seconds later. Although designed with three parachutes, two opening successfully is acceptable for the test perimeters and crew safety.”
However Boeing’s post-test press release makes no mention whatsoever of the parachute failure. Its a good thing that Jim Bridnestine directed Boeing to televise the test – otherwise we might not have known about the chute failure.
Keith’s second update: Boeing posted this update “Boeing statement regarding CST-100 Starliner pad abort test” saying “We will review the data to determine how all of the systems performed, including the parachute deployment sequence. We did have a deployment anomaly, not a parachute failure.” This is typical aerospace post-event mumbo jumbo. No one knows what happened so it is called an “anomaly”. I get that. But the parachute failed to deploy. We could all see that it failed to deploy. This update was not emailed to the same distribution list Boeing uses for press releases. Also, the earlier press release (that makes no mention of any parachute issues) is still online at Boeing. Anyone who sees this press release or the version sent out to the media may be totally unaware that the parachute failed to deploy on a vehicle designed to carry people.

NASA Watch founder, Explorers Club Fellow, ex-NASA, Away Teams, Journalist, Space & Astrobiology, Lapsed climber.

70 responses to “Boeing Starliner Pad Abort Test Was Technically A Success – But …”

  1. Bill Housley says:
    0
    0

    The commentary was incorrect. This is a nominal abort if there was a crew aboard. They test in order to shake out problems and they found a big problem that could have killed a crew, maybe. If root cause analysis finds that the other chutes opened because of good luck, then we can expect more delays.

    Also, is this the same chute system used on Dragon and Orion? Inquiring minds want to know.

    • Jack says:
      0
      0

      I doubt they have the same chute system unless they all weight the same which is unlikely.

      • Terry Stetler says:
        0
        0

        AIUI, Airborne Systems makes the chutes for Orion, Dragon and Starliner.

        One thing NASA found is that the modeling for asymmetry factor, the line loads in parachute arrays (n>1), weren’t accurately predicting it during a canopy deployment failure. Getting data for better models has been why so many tests have been done.

        SpaceX strengthened their Mk-3 parachutes by using Zylon in high load areas (per NASA/SpaceX newser 10/10/2019)

        • fcrary says:
          0
          0

          That’s a common manufacturer and probably a number of things in common with how they were designed. That could be a problem, but it’s not quite the same thing as them being the same design. (As an analogy, the Volkswagen emissions scandal affected about eight models, but a Beetle isn’t the same car as a Golf.)

          Those capsules have different masses (noted above) but also have different flight regime, affecting altitude and speed at deployment, and different numbers of parachutes. They aren’t going to be using the same parachutes.

    • rktsci says:
      0
      0

      There is a possibility that the same parachutes are used on all 3 – there aren’t many manufacturers of large parachutes. However, I expect that the deployment mechanisms are different on each capsule.

      • ThomasLMatula says:
        0
        0

        Which raises the question, did having to expand to support three different capsule systems result in a decline in their quality control and engineering expertise? Than you also have to factor in the 40 year long gap between the Apollo era and their developing the parachutes for the Cargo Dragon. Yes, they did have some work for planetary missions, but those were one off jobs designed for a very different environment. Than of course you have to factor in the extensive use today by engineers of simulations in place of hands on testing of hardware.

    • sunman42 says:
      0
      0

      Is it nominal if the number of successful deployments is two out of three? A normal understanding of variance would regard that as 2±1, which is a lot different than 3±1.

      • Bill Housley says:
        0
        0

        If the live commentary is to be trusted, then NASA and Boeing both seem to have said that the test was nominal in spite of the failed parachute deploy. They also seemed to say that 2 out of three chutes was within the parameters of that test…which kind of bugs this armchair quarterback. If I were to attempt to explain it I could only guess that they we’re testing the launch abort system, not the parachutes so it is somehow ok. Either that or that that overall (and I do mean “overall” in the broadest possible sense) record of of reliability of the parachute deployment system is way better than the redundancy used to protect against mission failure. Again, non-expert guessing on my part.

        • fcrary says:
          0
          0

          It’s entirely possible the criteria for a successful test was all about the abort systems and not the parachutes. That might even be a sensible way to write the test plan: If something extraneous were to go wrong, they wouldn’t be required to repeat the abort test.

          But that’s in the official world of checking off boxes and verifying formal requirements are satisfied. In parallel to that, there is (or should be) a process for resolving in-flight anomalies. If certain things go wrong, you aren’t supposed to fly again until you’ve figured out why and resolved the problem. That’s also true on aviation (although it grounds the specific vehicle, not the fleet) and even for your car (you can get a ticket for driving with a broken turn signal.)

          I’m reasonably sure a parachute on a capsule designed for astronauts would be on that list of critical equipment. The requirement to resolve such an anomaly can be waived. But NASA’s done that for O ring burn throughs and external tank foam shedding. They may be a little shy of doing that again.

          • Bill Housley says:
            0
            0

            Ya, I am quite sure that, internally, both NASA and Boeing take this chute deployment failure thing very seriously, even though they seem to ho-hum it externally.

            Just the same I’m very much looking forward to the root cause NOT being something that will delay Starliner significantly or Dragon or Orion at all.

            Actually, Orion is probably fine to get pushed down the calendar a little bit, since it’s waiting for it’s ride. However, Dragon and Starliner have a hard deadline to meet and can brook no further delays.

            Ironically, if the root cause is common between Dragon and Starliner, it is Dragon that will see the next delay since its in-flight abort is the next test and coming up fast.

  2. ThomasLMatula says:
    0
    0

    This is what comes from cutting corners by replacing testing with simulations. Boeing appears to have only done 5-6 parachute tests compared to the two dozen or so SpaceX has done. SpaceX reports the last dozen parachute tests were successful. Hopefully NASA will have Boeing go to SpaceX to get help with its parachute problems.

    It is good that the NASA Administrator forced Boeing to broadcast this test so the parachute failure isn’t hidden as it has been in past tests with only cryptic remarks about a “parachute” issue. A failed parachute is a pretty big issue in testing. They will need to fix this before a crew is sent up on Starliner.

    • Bill Housley says:
      0
      0

      It’s almost as if the post-test press release was written before the test.

      • fcrary says:
        0
        0

        That’s not uncommon. It’s best to get a press release out fast, while the news is still, well, new. So people write the press release in advance and assuming things went well. If they don’t, they can take a little extra time to rewrite it. Actually, if they are smart, they have several versions written in advance. There is precedent for this sort of thing. In advance of the D Day landings, Gen. Eisenhower had drafted two statements, announcing either success or failure.

        • kcowing says:
          0
          0

          So why is it still online?

          • fcrary says:
            0
            0

            I was simply saying that pre-written press releases are not uncommon. They are common, that is not unprecedented and there are sensible reasons for that practice.

            It also means that those pre-written press releases will be vague and need later follow up. Perhaps a retraction, perhaps a later, more detailed or accurate press release. Boeing has yet to do so, and that’s a reasonable basis for criticism. But when it comes to why they haven’t, I don’t think we have enough information to say. My personal preference is to blame stupidity rather than malice. But I can imagine a dozen possibilities between those two.

          • Bill Housley says:
            0
            0

            I know. I’ve done press releases and I know how they work.
            I know that they are prewritten, I sometimes do the same with my blog surrounding an event. I did it with the first Falcon Heavy launch…but when I do that I plan to edit it before release.
            There was an opportunity to edit this one, as others in the industry have surrounding these fickle and sometimes unpredictable rocket science things. Their spokesperson and the NASA rep acknowledged that the capsule was a chute short, so should the press release have done likewise.

    • Michael Spencer says:
      0
      0

      An open question, posed by a seeker of truth:

      Is this sentence:
      “SpaceX reports the last dozen parachute tests were successful”

      More accurately rephrased as:
      “In the last dozen tests, SpaceX reports that the parachutes deployed as expected”??

      Is this pedantic on my part?

      • ThomasLMatula says:
        0
        0

        Perhaps. The bottom line is that unlike Boeing parachutes, the SpaceX Mk3 parachutes worked.

        https://www.teslarati.com/s

        SpaceX says Crew Dragon parachute upgrade nailed more than a dozen tests in a row

        By Eric Ralph
        Posted on November 3, 2019

        • fcrary says:
          0
          0

          space.com is reporting 13 in a row, including one where they faulted one of the parachutes and landed within parameters on the other three. (Note that that was the test they previously had problems with.)

    • John Thomas says:
      0
      0

      How many tests had SpaceX performed before they had a parachute problem requiring more tests? Were SpaceX parachute issues related to ejection or opening up?

      • ThomasLMatula says:
        0
        0

        Good questions. Inquiring minds would like to see what the NASA reports actually said was the issue, versus the second hand information that has been released. But perhaps ITAR is involved.

      • Terry Stetler says:
        0
        0

        I don’t remember any SpaceX chute failures before NASA “requested” them to use 4 parachutes. This occurred between the 2015 pad abort test which used 3 parachutes and August 2016 when a test with 4 parachutes was reported.

    • Dewey Vanderhoff says:
      0
      0

      SpaceX uses an outside contractor for its parachute systems , Airborne Systems. And in turn that company relies on another outfit for the ” reel” system that works the all important chute lines and rigging. Boeing’s Starliner and Lockheed’s Orion use those same providers

      • ThomasLMatula says:
        0
        0

        Which would indicate that perhaps the SpaceX solution would work for the others as well. But also note this is why SpaceX really doesn’t like subcontracting work out. I suspect if SpaceX planned to use parachutes in future they would bring it in house, but since the Starship won’t be depending on parachutes for landing its probably not worth the hassle for a short production run system like the Dragon2.

        • Dewey Vanderhoff says:
          0
          0

          Airborne Systems has been around a century , celebrating its centennial this year. They do the abundance of military parachutes, everything from one man rigs and ejection seats all the way up thru heavy cargo chutes for airdropping , for all the armed services. I presume they have been working spaceflight since Sputnik as well . And now the space cadets. https://airborne-sys.com/home/

          • ThomasLMatula says:
            0
            0

            Than one has to wonder just what the problem is if they are so experienced. Also there never seemed to be any problem with the Cargo Dragon.

          • fcrary says:
            0
            0

            I have no idea about Airborne Systems’ experience with spaceflight, their relations with SpaceX or any personal experience with parachutes. But something did occur to me. In the past, didn’t packing a parachute take the sort of experience and practice which are hard to put into writing? Is that still the case? If so, isn’t that the sort of “craftsmanship” which has caused problems for other spaceflight projects? There is a certain fixation within the aerospace community for written procedures and doing things the same way every time (almost the definition of formal quality control.) That may not work so well for tasks which can be described as more of an art than an easily documented process. It might even hurt, if the artist needs to do something different for reasons he can’t explain and which aren’t in the manual.

          • Steve Pemberton says:
            0
            0

            Cargo chutes are typically subsonic and relatively low altitude. High altitude, high speed ejection chutes carry a light load. I wonder how much recent experience they have deploying three to four high altitude, high speed parachutes carrying a heavy load.

          • Dewey Vanderhoff says:
            0
            0

            I’m postulating that since their parachute expertise goes back to the days of the Sopwith Camel, and they seem to be the Go To shop for such things, it’s likely they may have provided chutes for Mercury, Gemini, Apollo, various LEO return capsules and the probes we sent to Mars. Somebody who knows should chime in.

          • Steve Pemberton says:
            0
            0

            I am wondering if they have done anything similar after Apollo. B-1A had an ejection module which would be somewhat similar to capsule reentry, not sure if they were involved with that but that also was a long time ago. F-111 had an ejection module and is a little more recent.

            If they have been involved with cargo Dragon then yes that is recent experience, but not decades long experience which is what people are referring to. Decades of experience with parachutes should not be discounted, but some comments in this and previous threads seem surprised why there should be any problems, what’s so new about parachutes. I think this is in a dynamic environment that even small changes in capsule design and flight profile quickly takes it into less well charted territory, requiring a lot more design and research efforts than many people were expecting. Interesting quote from Musk in September:

            “The Crew Dragon parachutes are way more difficult than they may seem. The Apollo program found them to be so hard that it became a notable morale problem!”

          • Todd Austin says:
            0
            0

            The Gemini and Apollo parachutes were made by the Ventura division of Northrup. The Mercury ‘chutes were made by the Radioplane division, also of Northrup.

            http://www.spaceaholic.com/

            https://airandspace.si.edu/

            https://airandspace.si.edu/

  3. ghall says:
    0
    0

    Four cameras and not one could keep the capsule in frame.

    • ThomasLMatula says:
      0
      0

      NASA said Boeing had to stream it live. NASA didn’t say the video had to be good. Hopefully there were other cameras filming it that had better, more consistent views.

      • TiminSoCal says:
        0
        0

        The cynic in me wonders if they thought “well we have to stream it, but it doesn’t have to be in focus or good at tracking the capsule”.

    • Steve Pemberton says:
      0
      0

      Or keep things in focus half of the time, I suspect some of the cameras were in autofocus which you shouldn’t do on launches because the flames throw it off. The whole thing seemed as if the camera operators and director had never shot a launch before. Or maybe they had but they didn’t adequately plan for how the abort test would be different from a launch.

      Made worse by not showing replays from all camera angles as is normally done, they just kept repeating the same original shot selections. Not that any of the angles were great but if we could have seen each angle in its entirety we could at least get a better idea of what happened. I wonder if the failure led to a decision to not replay all of the angles.

      • Bill Housley says:
        0
        0

        Come on. They ran the flight at the U.S. Army’s White Sands Missile range (at least I think I remember it belongs to the Army). it is reasonable to assume that somebody there knows how to record a rocket launch.

    • chuckc192000 says:
      0
      0

      It was almost like live streaming from cell phones.

  4. rb1957 says:
    0
    0

    The problem needs to be investigated and solved, but the test is successful if the landing was “survivable”. I’d’ve thought the abort system spec would say something (in much more complicated English) like …
    1) quickly remove the capsule from the danger, in a survivable manner; and
    2) safely land the capsule, in a survivable manner.

    • fcrary says:
      0
      0

      It all depends on the requirements and the test procedures. There probably is a high level requirement for probability of a survivable landing. And I also note that those parachutes are the same ones used for a nominal landing following reentry, so this anomaly is relevant to more than just an abort. That requirement would flow down to things like the probability of successful deployments, and perhaps having three parachutes to have an satisfactory probability of the necessary two deploying.

      So the failure of one parachute to deploy might have implications for the risk of _two_ to fail to deploy. Or maybe that possibility has been addressed in some other manner. We don’t have visibility into that level of detail. We do know that they are supposed to be designed for a less than one in 270 risk of failure for any reason. That’s not easy, and the parachutes are only one source of risk. Even if it’s a single event, it’s out of a relatively small number of tests. That can really mess up the statistics required to demonstrate that sort of reliability.

  5. chuckc192000 says:
    0
    0

    What bothered me more was the (toxic?) orange gas leaking from the crashed service module and its close proximity to where the Starliner landed.

    • Zed_WEASEL says:
      0
      0

      Not an issue with an actual launch from pad LC-41 at CCAFS. The service module will crashed into the ocean. However that means the Starliner is in the ocean as well.

    • Bill Housley says:
      0
      0

      One of the news articles noted that the toxic cloud actually passed over the capsule. So if this had been an actual failed launch pad explosion in the desert with crew aboard (never), then the rescue zone would have been contaminated. Kind of funny actually. An impossible failure more.

      I wasn’t flinging sand at the fact that SpaceX used a subcontractor. I think that was prudent in this situation by reason of that contractor being a highly experienced and trusted specialist. I’m just a little bit worried that the failure will be traced back to a common component and/or load procedure in all three of them…delaying all three of them further. The Demo 2 Orion and Demo 2 Crew Dragon have both already been delivered, correct?

  6. Winner says:
    0
    0

    So the failure of the primary third chute, leading to relying on only the minimial 2 required for a successful landing, is somehow a test success?
    That would mean that a 737 engine test flight would be successful if an engine fails, since you can land with one.

    • fcrary says:
      0
      0

      Not necessarily. That’s why people are supposed to write test procedures in advance and include success criteria. Rocket Labs lost their first launch of an Electron, and could shrug it off because they were (as they named the flight) “Just Testing.” Many tests involve stressing a system until it fails, to see how far you can push it. The part will fail, but the test will be a success.

      But in this case, I can’t see it. If their test procedure said, “Success is defined as having the capsule land with accelerations under X, no matter how that happens,” then I guess they could claim a successful test. But I really have trouble imagining anyone signing off to that. If the test procedure said, “demonstrate that the parachutes deploy correctly,” then “well, it worked anyway,” won’t do.

  7. Bad Horse says:
    0
    0

    Meh.. ? remember Apollo 15

  8. dbooker says:
    0
    0

    “Two of three Starliner’s main parachutes deployed just under half a minute into the test, and the service module separated from the crew module a few seconds later. Although designed with three parachutes, two opening successfully is acceptable for the test perimeters and crew safety.”
    And from NASA the day before Columbia launch,
    “Yes we’ve seen damage on the shuttle tiles due to foam liberation but all missions ended nominally…”

    And didn’t Boeing insist MCAS was safe?

    Just saying…

  9. james w barnard says:
    0
    0

    Until Boeing stated that the failure of the third chute to deploy was an “anomaly” I was thinking perhaps this was an intentional non-deployment to gain data on the loads a crew would encounter in a “single-chute-out” situation. Such a test would make sense. I don’t recall if we did test drops of the mass-simulators on the Shuttle SRB parachute recover system with one of the 136ft diameter main chutes out, but we successfully recovered at least one SRB with a single failed chute, although damage to the aft skirt was more extensive on impact in the ocean.

  10. Shaw_Bob says:
    0
    0

    Two parachutes instead of three is NOT a success. Nobody should fly on this spacecraft until it demonstrates success. And nobody should fly on any first orbital test, either.

    • Mark says:
      0
      0

      No one is going to fly on the first orbital test.

      • kcowing says:
        0
        0

        But people will fly on the second orbital flight using systems identical to the ones that did not perform today.

        • fcrary says:
          0
          0

          Let’s wait for more information. Boeing may make changes based on this event, and either NASA or the ASAP may insist on changes. It isn’t certain that astronauts will fly with exactly the same parachute system used in this test. One reason for tests is, after all, to identify problems and figure out how to fix them. Whether or not that will happen is an open question, but let’s not rule out the possibility.

        • Mark says:
          0
          0

          And the same contractor makes similar systems for both Dragon and Starliner and we don’t know how much is of the two systems is interchangeable and if it was a issue with an interchangeable part or a part unique to Starliner’s chute.

    • Bill Housley says:
      0
      0

      The pad abort test, of systems that all of us hope will never have to be used, was a complete success. NASA and Boeing had apparently already defined this test such that the failure of one chute would still be a successful test…no doubt because it wasn’t he chutes that were being tested. However, that successful test has now uncovered a failure mode in a routine flight system that, between Starliner, Dragon, and Orion, will probably be used at least a hundred times in the lives of those programs. This failure mode might even be common to all three spacecraft.
      Even though the planned redundancy is overkill to the demonstrated odds of failure, they will now still need to dig in, find, and fix it and that might cause further delays in this, and perhaps all three, already late programs.

  11. jimlux says:
    0
    0

    Where did the big cloud of Nitrogen Tetroxide come from? Service module tanks rupturing on impact? It was off screen to the right, but that was an impressively big fluffy red cloud.

    • kcowing says:
      0
      0

      Yea we used to call it MFRG

    • Zed_WEASEL says:
      0
      0

      AIUI the Starliner service module was venting NTO after the abort to prevent ignition of the hypergolic propellants from the lithobraking event.

      • Jeff2Space says:
        0
        0

        Sure, but why is it venting corrosive nitrogen tetroxide while deploying parachutes? That seems… bad.

        I would think you’d want to vent the service module well after it’s gotten away from the capsule, but before it hits the ground/ocean.

  12. Winner says:
    0
    0

    Boeing, the company that is making “an already safe plane even safer” (737 MAX), is posting the “alternative facts” in the primary press release that don’t need to mention the pesky details you cite in your blurb. Thanks for keeping us informed Keith/NASAWatch.

  13. Bill Housley says:
    0
    0

    Keith,

    On your most recent update: I get accused of not really knowing what I’m talking about because I’m just scifi guy, I’m not a rocket scientist, I don’t have a college degree, and I’ve never worked in the industry. It’s comforting when my initial impressions match yours and/or the other experts on this panel.
    It looks to me like Boeing doesn’t want this to impact their reputation, and it shouldn’t, but there has got to be a less slimy-looking way to achieve that. It’s not like nobody noticed, everybody did, and pretending like it’s not a big deal reduces confidence in them more than they deserve…not less.

    • TiminSoCal says:
      0
      0

      Boeing hasn’t been real good at owning their mistakes lately, in Space as well as commercial aviation.

  14. Tom Mazowiesky says:
    0
    0

    Were any astronaut personnel at the test? Any comments from them on the ‘anomaly’?

    Hopefully after the MCAS mess, Boeing will be more forthright on this problem. They will need some time to examine the flight data, the visual record, etc., and come up with an analysis of what caused the problem and what fixes may be needed to prevent it from happening again.

    In addition to the deployment problem, what was up with the flames that seemed to envelop part of the spacecraft? Was this nominal (yikes!)?

    • Steve Pemberton says:
      0
      0

      There was some pretty bright sunlight reflections off of the capsule in some of the shots, exacerbated by being out of focus. So depending on what part of the test you are referring to that may have been what you were seeing.

      All three crewmembers for CFT were there, they appeared afterwards for an interview and seemed upbeat. Naturally they aren’t going to say anything negative, but I’m guessing they realize it just means there’s still a ways to go before they get to fly their mission.

      • Terry Stetler says:
        0
        0

        Sunlight reflecting? Through that cloud??

        https://twitter.com/FxPhilW

        • Steve Pemberton says:
          0
          0

          I was replying to a comment “what was up with the flames that seemed to envelop part of the spacecraft”. As I said I’m not sure what they were looking at, I suspected they might be referring to the bright yellow that lasts for several seconds, not the small reddish area that the screenshot that you posted shows which is only seen by studying frame by frame as the reddish area is only seen for a few milliseconds.

          The bright yellow area seems to be sun reflection in one camera angle, exacerbated by that camera being out of focus. I base my opinion on the fact that it is only seen in one camera angle. Again made difficult since they did not replay each camera angle in full, but in the multi-angle view the yellow brightening is seen only in one of the four camera angles, then they cut to that angle exclusively for several seconds (when the screenshot you showed was taken) then they cut to a different angle where the capsule looks normal.

          The reddish area yes that looks like potential flame, it’s very brief but of course it doesn’t take long to do damage. If it does come out that burning fuel caused one of the parachutes to fail that is not something Boeing will be able to spin away as being acceptable (hey, the fire only burned off one chute so that’s okay). Of course this is all speculation until a report on this (hopefully) comes out.

          • Tom Mazowiesky says:
            0
            0

            Could be a reflection, I looked at the video a couple of more times and it is equivocal, could be a sun flare or fire or a bit of both. Examination of the spacecraft and the chute rigging may show what actually occurred. As you say, we’ll just have to wait for the formal reports.

      • Tom Mazowiesky says:
        0
        0

        I think a sunlight flare would cause reflections off the internal optics to appear, you see that on pretty much any lens with multiple elements. Sure looks like fire to me.

  15. ed2291 says:
    0
    0

    This illustrates a clear difference between Boeing and Space X. Both are equally prone to failure, but Space X’s clear and ambitious goals as well as transparency about a failure and what they are doing to correct it give Space X more credibility.

  16. Michael Spencer says:
    0
    0

    oh, never mind.