Failed Strut Doomed CRS-7 Mission; No Falcon 9 Launches Before September, Says Elon Musk

The nine plumes of the Merlin 1D engines flare as the vehicle disintegrates. Photo Credit: Alan Walters/AmericaSpace
The nine plumes of the Merlin 1D engines flare as the vehicle disintegrates. Photo Credit: Alan Walters/AmericaSpace

Three weeks after the catastrophic loss of a Falcon 9 v1.1 booster, which exploded just 139 seconds after leaving Space Launch Complex (SLC)-40 at Cape Canaveral Air Force Station, Fla.—dooming the seventh dedicated Dragon cargo ship, bound for the International Space Station (ISS)—SpaceX CEO Elon Musk described the status of his company’s ongoing investigation on Monday, 20 July, and revealed that a failed helium tank strut appeared to be the root cause. Over the past weeks, engineering teams have spent thousands of hours matching up data across Falcon 9 v1.1 systems down to the millisecond, in order to properly understand the event which befell their vehicle, within a period of just 0.893 seconds, prior to loss of telemetry. In summarizing the developments, Mr. Musk noted that the accident occurred rapidly, that software modifications might have saved its precious Dragon payload, and that the failure is not expected to impair SpaceX’s goal of launching its first Commercial Crew missions from 2017 onward. At the same time, he added that the Hawthorne, Calif.-based launch services company does not expect to fly another Falcon 9 v1.1 until September 2015, at the soonest.

As described in a previous AmericaSpace article, the loss of the CRS-7 mission on 28 June shattered an impressive track record of 18 successful launches by SpaceX since June 2010, which have delivered a wide range of payloads into low-Earth orbit, Geostationary Transfer Orbit (GTO), and further afield, to the L2 Lagrange Point. Thirteen successful missions have been flown by the upgraded Falcon 9 v1.1 since September 2013. This variant of the two-stage booster benefits from the enhanced Merlin 1D powerplant—whose nine first-stage engines are capable of 1.3 million pounds (590,000 kg) of propulsive yield and whose restartable Merlin 1D Vacuum second-stage engine generates 180,000 pounds (81,600 kg) of thrust—and can deliver up to 28,990 pounds (13,150 kg) into low-Earth orbit and up to 10,690 pounds (4,850 kg) into GTO. Key payloads launched successfully to date by the v1.1 have included four ISS-bound Dragons, the Deep Space Climate Observatory (DSCOVR) to the L2 Lagrange Point, and six telecommunications satellites into Geostationary Transfer Orbits at an approximate altitude of 22,300 miles (35,600 km).

First-stage flight of the CRS-7 mission appeared nominal, with catastrophe striking the Falcon 9 v1.1 abruptly. Photo Credit: Alan Walters/AmericaSpace
First-stage flight of the CRS-7 mission appeared nominal, with catastrophe striking the Falcon 9 v1.1 abruptly. Photo Credit: Alan Walters/AmericaSpace

However, the heart was torn from SpaceX on 28 June, when the 14th Falcon 9 v1.1 vanished into a fireball, just 139 seconds after liftoff, dooming not only the CRS-7 Dragon cargo ship—the seventh dedicated mission under the terms of a $1.6 billion Commercial Resupply Services contract with NASA, signed back in December 2008—but also the first of two critical International Docking Adapters (IDA-1). These adapters, the second of which remains earmarked to fly aboard the CRS-9 Dragon mission, were intended to be installed onto the forward-facing and space-facing (or “zenith”) ports of the space station’s Harmony node, thereby providing primary and backup docking interfaces for Boeing’s CST-100 and SpaceX’s own Dragon V-2 piloted vehicles from mid-2017 onward. It is understood that IDA-2 will now be repurposed for the former IDA-1 role, whilst a set of structural spares will be used as the basis for the construction of IDA-3 to fulfil the original IDA-2 role. Although it is expected that IDA-3 will ride a Falcon 9 v1.1 uphill, it remains to be seen when that launch will now take place.

Speaking earlier today, in a media teleconference, SpaceX CEO Elon Musk explained that a support strut—measuring about 2 feet (60 cm) in length and about an inch (2.5 cm) in thickness at its widest point—appeared to be the root cause of the failure. Developed by a supplier which Mr. Musk refused to name, the strut’s role was to support one of the composite over-wrapped helium tanks in the Falcon 9 v1.1’s second stage. Designed to handle up to 10,000 pounds (4,530 kg) of force, the strut actually failed at just 2,000 pounds (900 kg), barely a fifth of its rated strength, catastrophically releasing helium into the second stage Liquid Oxygen (LOX) tank at about 3.2 G. “The pressurization system itself was performing nominally, but with the failure of this strut, the helium system integrity was breached,” SpaceX explained in a news release Monday evening. “This caused a high-pressure event inside the second stage, within less than one second, and the stage was no longer able to maintain its structural integrity.” It is understood that the struts will no longer be used on future missions and that SpaceX plans to “implement additional hardware quality audits throughout the vehicle” to ensure that “all parts received perform as expected per their certification documentation.”

Subsequent inspection of pre-flight images of the strut revealed no apparent trace of damage. “No evidence of assembly errors in the strut in hi-res closeout photos taken before launch,” Mr. Musk stressed, adding that SpaceX engineers will now move toward individual testing of the struts. He noted that the accident unfolded rapidly, with the first indication of a problem through to the structural breakup spanning just 0.893 seconds, and described the circumstances of the CRS-7 accident as a “really odd failure mode.”

Within a matter of seconds, the Falcon 9 v1.1 was reduced to debris, plunging back to Earth. Photo Credit: Alan Walters/AmericaSpace
Within a matter of seconds, the Falcon 9 v1.1 was reduced to debris, plunging back to Earth. Photo Credit: Alan Walters/AmericaSpace

These findings are in keeping with early speculation on 28 June, which highlighted that first-stage performance was entirely nominal and that the CRS-7 Dragon spacecraft continued to communicate, until it passed over the horizon after the failure. Mr. Musk pointed out that, in spite of no other issues with the launch, SpaceX was looking at other issues, admitting that the company may have become complacent in recent years. “Rockets are a fundamentally difficult thing,” he explained, adding that no further Falcon 9 v1.1 launches are anticipated before September 2015 at the earliest, with the maiden flight of the mammoth Falcon Heavy—which, when operational, carries the potential to deliver up to 117,000 pounds (53,000 kg) into low-Earth orbit and up to 46,700 pounds (21,200 kg) into Geostationary Transfer Orbit—pushed back to no sooner than March 2016.

This is an intense pity, for SpaceX accomplished five flawless Falcon 9 v1.1 missions in the first four months of 2015, together with the highly successful Pad Abort Test (PAT) of its crewed Dragon vehicle in May. The company has delivered a pair of Dragons—CRS-5 in January and CRS-6 in April—to restock the ISS, despatched the DSCOVR spacecraft to the L2 Lagrange Point in February, and lofted the Eutelsat 115 West B, ABS-3A and TurkmenÄlem52E/MonacoSat communications satellites into Geostationary Transfer Orbit in March and April. Had the loss of CRS-7 not occurred, as many as eight more missions were timetabled for the second half of 2015, including a further two Dragons to deliver the Bigelow Expandable Activity Module (BEAM) and the IDA-2 docking adapter to the space station, together with the Jason-3 ocean surface altimetry mission, several commercial payloads, and the maiden voyage of the Falcon Heavy.

In spite of the disappointment, Mr. Musk does not expect the CRS-7 loss to significantly impact the Commercial Crew timeline, describing it as being “not on the critical path,” and kindling some hope that SpaceX may still be in a position to launch its first uncrewed Dragon as soon as the fall of 2016 and a piloted mission several months later. He added that had the “right software” been aboard CRS-7, the Dragon and its payload might have been salvaged. “The Dragon spacecraft not only survived the second-stage event,” a subsequent SpaceX news release stated, “but also continued to communicate until the vehicle dropped below the horizon and out of range.”

Software to permit the deployment of parachutes in the event of a launch failure will be included aboard the CRS-8 Dragon mission, which was originally scheduled for early September, but whose target date presently remains in flux. Although Mr. Musk hinted at a No Earlier Than (NET) target of September for the next Falcon 9 v1.1 mission—with no information yet on whether it will carry Jason-3, CRS-8 or another payload—he was adamant that precise return-to-flight dates would only be issued after all data had been fully analyzed and properly understood. What is clear, though, is that lost revenues from this failure are expected to be “meaningful,” stretching into the hundreds of millions of dollars. “Our investigation is ongoing until we exonerate all other aspects of the vehicle,” SpaceX stressed, “but at this time, we expect to return to flight this fall and fly all the customers we intended to fly in 2015 by end of year.”

 

The author would like to express thanks to AmericaSpace’s Michael Galindo, for his support in the preparation of this article.

 

 

Want to keep up-to-date with all things space? Be sure to “Like” AmericaSpace on Facebook and follow us on Twitter: @AmericaSpace

34 Comments

  1. Hi Ben,

    Excellent factual article as usual.

    If I might, a question.

    Several other articles have implied that SpaceX did not do acceptance testing on the struts:

    “Before this, the company had trusted that the manufacturer — which Musk won’t name — was providing sturdy struts.”

    http://www.businessinsider.com/why-spacex-rocket-exploded-2015-7

    And was not even requiring certification testing documentation, only materials certification.

    Any idea if this is true?

    Thanks.

      • Tracy,

        If you are kidding great.

        If you are seriously making accusations of sabotage, please make specific accusations (who, how, why) and provide supporting evidence.

        • Joe,
          My thoughts as to the answers of your question.

          The How

          “Speaking earlier today, in a media teleconference, SpaceX CEO Elon Musk explained that a support strut—measuring about 2 feet (60 cm) in length and about an inch (2.5 cm) in thickness at its widest point—appeared to be the root cause of the failure. Developed by a supplier which Mr. Musk refused to name, the strut’s role was to support one of the composite over-wrapped helium tanks in the Falcon 9 v1.1’s second stage. Designed to handle up to 10,000 pounds (4,530 kg) of force, the strut actually failed at just 2,000 pounds (900 kg), barely a fifth of its rated strength, catastrophically releasing helium into the second stage Liquid Oxygen (LOX) tank at about 3.2 G.”
          This equipment was located in the unpressurized area of the 2nd stage and represents a much easier access point for the purposes of sabotage. How does a strut fail at 20% of design load? It doesn’t unless somebody wanted it too.

          The Why

          SpaceX competes with participants whose cash flows are adversely affected or potentially so from the continued market disruption that SpaceX represents. SpaceX is presently 25% to 30% less in cost than all other launch providers. The introduction of the reusing the first stage will lower costs potentially another 25% to 30% again. This will create a margin of 50% to 60% difference in launch costs reductions. This will further impact the heavy lift market that SLS will be included in. This market disruption has been met with annouced reusable systems by Lockheed Martin, Boeing, BAE and others. It was considered by many even on this Website that SpaceX had a small chance of success with such a system. Yet they have shown real progress in landing the first stage on a barge.

          The Who

          This list in not any particular order. Airbus, Lockheed Martin, Boeing, United Technologies, General Dynamics, BAE, Northrup Grumman, Raytheon, General Electric, China, Russia, Blue Origin, Virgin Galactic, Orbital Sciences.

          The Result

          SpaceX reusability program has been delayed indefinitely. The BEAM prototype for Bigelow has been delayed indefinitely. Russia is in negotiations with NASA to provide crew transport to the ISS beyond the 2017 current contract because both SpaceX and Orbital launchers have experienced system failures that are part of the crew launch systems. This failure provides future funds to Russia, maintains cash flow to current operators and reduces market disruption significantly.

          • I was following you on the potential motives (not that I think sabotage at all likely – as there so far appears to be zero evidence of it) but the Result doesn’t seem to mesh with what’s been reported in the media.

            Reusability program delayed indefinitely – Nope – They’re talking about a return to flight this year, possibly as early as September.

            BEAM prototype delayed indefinitely – Nope – Just not rescheduled yet, and getting BEAM proven sooner would benefit Lockheed Martin, Boeing and ULA so it takes them off the “suspect” list, since Boeing is partnered with Bigelow on CST-100 launched on an Atlas V to take people to the station Bigelow wants to fly – scratch Roscosmos and Airbus from the list too, since their competitors to win launch contracts for Bigelow’s modules.
            Russia negotiating for crew launches – Nope – As announced in the conference, flight delays are not expected to impact commercial crew, and Orbital’s launch failure has nothing to do with Commercial Crew, the Antares was not and won’t become human rated. Bolden has already stated they still intend to book seats on Soyuz when/if Commercial Crew is in operation, with Russia booking seats on Commercial Crew flights in return, as was done during the Shuttle era.

            And what about the means? Covertly controlling a sub-contractor’s factory to be able to produce a part with a great enough flaw to cause a failure without being detectable by both the manufacturer’s QA or SpaceX incoming inspection policies doesn’t sound like typical mob operation to me.

            Deep conspiracy, or what it looks like on face value… A gap in procedure that allowed an uncommon but critical flaw to get through into the finished product.

            • Tim,
              I am not sure if there ever has been a public admission that sabotage has been ever listed as the cause of a launch failure…Are you aware of this ever occurring within the industry? Other than the Russians who always blame sabotage on their rocket failures.

          • Tracy,

            You say:

            “This equipment was located in the unpressurized area of the 2nd stage and represents a much easier access point for the purposes of sabotage. How does a strut fail at 20% of design load? It doesn’t unless somebody wanted it too.”

            My understanding is that the strut in question was inside the second stage LOX tank, not an easy access point.

            A much more likely reason for a strut to fail at 20% of design load is a badly designed and inadequately tested design.

    • The audio from the conference call where Musk described the preliminary results is available here: http://images.spaceref.com/audio/companies/spacex/2015/SpaceX_CRS7_Update_20150720.mp3

      I haven’t found a good complete transcription yet, but from my own listening, a couple of Musk’s statements stand out that I think are key to your question:

      ==
      07:05:
      It should be said however that we’ve flown several hundred such struts on every Falcon 9 vehicle, meaning there’s a cumulative flight history of several thousand of these struts. Moreover, the strut that we believe failed was designed and material certified to handle 10k lbf but actually failed at 2k lbf.

      09:28:
      What we’re going to do in the future is we’re obviously not going to use these particular struts
      and we’re going to move to individually testing each strut independent of any material certification, so no matter what something is actually certified to handle, we’re not going to believe that,
      we’re simply going to individually unit test every single strut.

      21:43:
      We were relying on material certification from the supplier and so (..) in the future we (..)

      independent material certification, we will be individually testing each strut.

      I think that strongly suggests that SpaceX was not doing its own testing on this component and also not requiring documentation of testing from the supplier.

      • There’s also a question of what kind of testing was being done both on the manufacturer’s end and SpaceX.

        If they tested 1,000 units to their failure point and 5 or 6 (a decent interpretation of “a few”) failed, that’s something they might never have seen if either the manufacturer or SpaceX was pulling 5 percent of the stock and destruction testing them as representative of each lot. It’s just a game of odds.

        It may sound easy for Musk to say “we will be individually testing every strut,” but a working solution isn’t that simple. Testing 100% of the incoming struts to see where they fail would leave them with a shop full of broken struts and nothing to build with. They’ll need to identify the cause of the failure (one tweet from a journalist quoting from the call referred to grain structure of the metal in the struts as suspect) and develop a non-destructive test for that.

      • Matt,

        Thanks for the information, it certainly seems to confirm what I heard.

        Tim,

        “Testing 100% of the incoming struts to see where they fail would leave them with a shop full of broken struts and nothing to build with.”

        Good point.

        Among my other bad habits I have been known to get paid for writing verification requirements. For flight critical hardware the least verification I have ever encountered was “Review of Lower Level Documentation”. That would involve design drawings, materials selection, and testing of the design by the provider. The units would only be accepted when (and if) the structures people signed off on it (and thus put there own necks on the line).

        Personally I would require acceptance testing by the receiver (SpaceX in this case) as well. That kind of thing can be expensive and the customer can overrule, but if they did; in a case like this the requirements writer is on record as having made the recommendation.

        If SpaceX cut these kind of corners here, the question becomes on what else did they cut corners.

        • Joe….,

          “Among my other bad habits I have been known to get paid for writing verification requirements. For flight critical hardware the least verification I have ever encountered was “Review of Lower Level Documentation”. That would involve design drawings, materials selection, and testing of the design by the provider. The units would only be accepted when (and if) the structures people signed off on it (and thus put there own necks on the line).”

          In your professional experience have you ever encountered such a blatant design and or material flaw within the supply chain? Would this be incompetence or negligence. Does this rise to the level of gross negligence?

          • Tracy,

            “Would this be incompetence or negligence. Does this rise to the level of gross negligence?”

            In my opinion (for whatever it is worth) there is not yet enough publically available information to reach such conclusions.

            There is, however, enough information to be extremely skeptical of what will happen next.

            I suspect that the DoD (in particular) are going to be very interested in how diligent SpaceX is in assuring the quality of its product. So will insurance companies asked to insure payloads that may be more expensive than the rocket on which they are launched.

            We will all have to wait to find out what the final resolution will be.

            But while we are waiting, consider the following:

            One of Musk’s assertions is that SpaceX will soon be reusing the Falcon 9 first stages (which apparently includes these suspect struts) as many as 1,000 times with minimal inspection and very short turnaround times.
            What would be the extra stress on such components on the second, tenth, one hundredth, one thousandth launch.
            How credible (at this point) are such plans?

            Believe it or not I do not like raining on other peoples parades, but sometimes a little rain may keep you from being struck by lightening.

            • So in other words as much as I would like this to be some elaborate scheme to stop market innovation…In reality its just incompetence and arrogance…Clearly we are doomed as a species..

              • “Clearly we are doomed as a species.”

                My assessment of the human race’s chance of survival is no where near that pessimistic.

                But assuming that the under performing strut is the sole cause of the CRS-7 failure (remember the investigation is still ongoing), then – yes – the cause of the incident was a substandard component that got past a deficient certification/acceptance process.

              • As a species we’ve survived the catastrophic failure of many rocket launches caused by parts that either failed to meet their design requirements and weren’t caught by effective QA, or for which the design requirements were not adequate to start with. Always it boils down to somewhere, someone wasn’t competent enough to catch a critical detail, and may well have been arrogant enough to think they wouldn’t miss something critical.

                I think we’ll survive this one, too.

  2. For want of a strut, the spacecraft was lost… almost sounds like a proverb… 😉

  3. “How credible (at this point) are such plans?”

    Considering that Musk has said that the Merlin engines would need rebuilding after 40 cycles, I doubt that over a thousand uses of an F9 first stage only with only minimal inspection is at all credible, but I also have yet to find where he actually claimed it would soon be possible.

    As for re-use at all and what inspection levels will be necessary – that really remains to be seen. NASA’s still not planning on anything but virgin boosters for manned spaceflight or even resupply missions. There are military aircraft that hit multi-mach speeds and high G-loads that manage to be re-used with appropriate materials and manufacturing QA and inspection/maintenance process.

    Time will tell what risks commercial customers are willing to take after the first few (if/when success is achieved) first stages are landed and then re-flown to destruction to get a handle on their practical lifespan.

    • “Considering that Musk has said that the Merlin engines would need rebuilding after 40 cycles, I doubt that over a thousand uses of an F9 first stage only with only minimal inspection is at all credible, but I also have yet to find where he actually claimed it would soon be possible.”

      A couple of points:

      I believe it was Gwynne Shotwell, not Musk, who talked about the Merlin Engine requirements.
      Unfortunately it seemed so strange that I did not save the links (about 1,000 reuses), but Musk said it and certainly a lot of SpaceX on-line supporters began touting it as a near future “reality”.
      In any case the real point is (assuming that it will be possible to reuse the stage at all) what will be the trades for reusing the stage 40 times or twice for that matter. Will the turnaround/maintenance costs be low enough to make such a plan less expensive than building a new expendable launcher? How much cheaper would either approach be than the “competition”?

      When SpaceX first began talking about recovering the first stage it was to study it to improve the design to make a more reliable expendable booster. At that point I was very supportive (Yes at one time I liked SpaceX). That somehow morphed in to the current situation.

      Not trying to “pick a fight”, just some things that should be considered with the hypothetical strut failure as a reality check.

      • I believe you are right about Shotwell throwing out the 40 cycle number. When doing the Reddit Ask Me Anything interview some months back, Musk was asked specifically about what that 40 cycle number meant to which he replied that there were wear items in the engine that would need to be replaced after 40 cycles.

        Yes – there are definitely rabid hordes of devotes who are convinced that we’re easily only 5 years away from rapid reusable rockets and a permanent colony on Mars. I know during the the unveiling of the powered landing concepts for the F9 and Dragon Musk talked about rapid re-use goals in the long term, but from articles around that time it seemed like his position wasn’t minimal inspection, but rather automating inspection to provide detailed inspection quickly. Even then, were SpaceX able to somehow magically crack that nut tomorrow – having a total of two launch pads, and two under construction, none of which are approved for more than 12 launches a year that Tomorrowland(tm) future of rapid re-use and high cyclic rates won’t be happening soon.

        I think there may be more to consider about re-use than price. Even if re-use could provide a marginal or negligible price break – considering SpaceX’ backlog it’s not unrealistic to see their launch rate of virgin cores exceeding their maximum production rate. Shotwell has already cited the reason for their consideration of fairing re-use not being cost-oriented but rather to deal with the bottle-neck of fairing production volume, with a recovery and refurbishment program possibly being more cost effective than setting up a second production line. Should SpaceX satellite division become a SpaceX launch customer I could easily see them taking the risk of launching on re-used first stages in order to not have to wait for virgin cores, especially with their goal of high volume low cost satellites – lower payload value would make them less risk averse.

        I also agree that with what info has surfaced so far it’s difficult to really understand all the ramifications of the strut failure (or if the accident reconstruction is accurate) but if it is as it appears – a materials quality issue, that doesn’t seem at all beyond something that can be addressed – whether reused or not. If a strut failed in one flight because its metal was not sound (porosity, grain alignment, etc.) it could fail after 3 flights, or 20 or whatever – but that doesn’t mean that struts meeting the spec are any more likely to fail after 20 flights than 1 (barring of course the discovery of some unknown repetitive stress that would be found during destructive high-volume repetitive flight testing). The aircraft industry deals with similar problems on reusable mach+ vehicles that take higher G loads from all directions, yet it’s often much less expensive to replace components based on predicted wear levels and known flight stress data, then to scrap the vehicles after each flight.

        Definitely not seeing that as trying to pig a fight. Some folks think that’s what any differing opinion on the Internet is – this looks more like discussion to me.

        • “Even if re-use could provide a marginal or negligible price break – considering SpaceX’ backlog it’s not unrealistic to see their launch rate of virgin cores exceeding their maximum production rate. Shotwell has already cited the reason for their consideration of fairing re-use not being cost-oriented but rather to deal with the bottle-neck of fairing production volume, with a recovery and refurbishment program possibly being more cost effective than setting up a second production line.”

          An interesting concept. I have a hard time believing that recovery/refurbishment/reuse of fairings (much less boosters) could really be more efficient than building new units, but I am open minded on the subject.

          “Should SpaceX satellite division become a SpaceX launch customer I could easily see them taking the risk of launching on re-used first stages in order to not have to wait for virgin cores, especially with their goal of high volume low cost satellites – lower payload value would make them less risk averse.”

          I assume this refers to the satellite internet concept, I thought Musk had recently been backing off any near term implementation of the concept.

          “The aircraft industry deals with similar problems on reusable mach+ vehicles that take higher G loads from all directions, yet it’s often much less expensive to replace components based on predicted wear levels and known flight stress data, then to scrap the vehicles after each flight.”

          Comparing maintenance of aircraft to spacecraft is always tricky. During the DC-X program in the 1990’s I had the chance to sit in on meetings between the late Dr. William Gaubatz and his designers about allowing for maintenance on a reusable rocket. I do not doubt it can be done, but have seen no evidence that such considerations have been taken into account in the design of the Falcon 9.

          • If the condition of fairings washing up on shores of the Bahamas, Carolinas and South Africa is an indicator – refurb doesn’t sound that impractical to me – they’re physically torn apart by waves, but not charred – GoPro cameras with plastic lens housings still intact and undamaged. I think the bigger issue would be cost of having helicopters on station to snag them as they parachute down. Surely a balancing act – what’s more practical – difficult recovery, or building and staffing another production line. Shotwell’s take at a satellite conference this spring was that the production line set-up cost is huge enough that recovery is worth investigating.

            I don’t think the satellite constellation will be soon either, but neither will reusable boosters. Even if they’re flying F9 again this year and get their first couple of boosters recovered within a year, that’s still a long way from testing and developing a re-use program.

            I expect the comparison to high performance fighter aircraft would be closer to an orbital launch rocket than the comparison Musk makes to airliners, with the supersonic speeds and heavy dynamic g-loading – and they are more maintenance intensive – but still maintenance wins out over replacement.

            I’d love to have been a fly on the wall for DC-X engineering meetings. That’s some good history. Isn’t that also a comparison to an aircraft too, albeit one that was doing research for a spacecraft?

            • “I think the bigger issue would be cost of having helicopters on station to snag them as they parachute down.”

              Agreed. Have had a repetitive conversation with SpaceX supporters about reusability. It is not desirable because it is cool (although, of course, it is);
              it is only desirable if it is in someway more efficient.

              Another point to consider is the weight of the parachutes and grappling system for fairing recovery. That would all have to be subtracted from the payload.

              “I’d love to have been a fly on the wall for DC-X engineering meetings. That’s some good history. Isn’t that also a comparison to an aircraft too, albeit one that was doing research for a spacecraft?”

              Yes. What the DC-X program had done was bring in aircraft maintenance personnel from the Long Beach Aircraft plant to work with the Huntington Beach Rocket personnel. The Monday morning meeting was to allow the two groups to present their differences to the Program Lead (Gaubatz) for resolution. The meetings could best be described as frank and intense. That eventually produced a maintenance friendly design concept (tested on the DC-X) to be used for the eventual Delta Clipper.

              The point I was getting at is that no such work was done (to the best of my knowledge) during the design phase for the Falcon 9. It was initially designed to be a reliable and (relatively) low cost expendable booster (a noble goal in itself) and has now morphed into a reusable rocket program. That means they would have to back drive maintenance requirements into the design. That will not be an easy trick.

              • That must have been amazing to listen in on.

                I’ve been under the impression that re-use was SpaceX plan from the get go. Long before the Falcon 9 flew and before the powered landing announcement they had announced plans for re-use via parachute recovery and splashdown. Certainly there’s a major design change from that to their current approach.

                That said, making it fly and making it practical to re-fly are far from the same thing. If they do succeed at re-use I wouldn’t be at all surprised to see another design iteration focused on making it more practical to inspect and or repair.

                • “That must have been amazing to listen in on.”

                  It was.

                  An interesting thing to keep track of will be what Blue Origin is doing (as there activities – hopefully – become more publicly known over the next couple of years).

                  Several of the Delta Clipper Team now work for Blue Origin and Dr. Gaubatz was a consultant to them until his untimely demise last year.

                  No way to know for sure, but they have probably been designing for maintenance from the beginning.

  4. I’ve read that since identifying the problem testing has shown too many of those struts do not meet requirements. 20% is a huge margin. They should have expected better from their supplier. It may become another part they do in house.

    Sabotage is possible but not likely. It doesn’t mean they shouldn’t take precautions.

    SpaceX continues to impress. Identifying this failure required understanding the interaction of multiple stream of sub second data from a destroyed vehicle (which tends to disrupt your data.) The only real mistake they seemed to have made is not anticipating the possibility so they would upgrade the capsule software which would have allowed it to be recovered. Had they done that we’d have to assume they were beyond human.

    • “I’ve read that since identifying the problem testing has shown too many of those struts do not meet requirements. 20% is a huge margin.”

      The 20% is not a margin. According to SpaceX the strut failed at about 20% of what was supposed to be its design load.

      “They should have expected better from their supplier. It may become another part they do in house.”

      That is not what they say. They say they will begin acceptance testing of the received parts, which they should have been doing in the first place.

      “SpaceX continues to impress. …Had they done that we’d have to assume they were beyond human.”

      I am sure they continue to (over) impress some and that you would like to believe they (meaning Musk) are beyond human.

      Others might wish to withhold judgment.

      • I am one impressed with performance to date, that has no problem withholding judgement until I see how this situation works out. I believe that reusing tested vehicles will gain in reliability, but am also willing to withhold judgement on SpaceX accomplishing it pending results.

        Having run a company for quite some time with myself being the indispensable man, I know the weaknesses of any enterprise dependent on one individual. If I make a certain types of mistake, we have a real problem. If I am not there personally in some situations, we have a real problem. Management depth could be a problem with SpaceX that shows up in unforeseen ways.

        There is a fuzzy boundary between cost cutting and quality control with some firmly in both extremes. IMO, excess in either direction results in an unworkable system. I think the EELVs are too far in one direction, and results of the current problem will show if SpaceX is too far in the other direction. I think 94% success on the Falcon9 to date suggests a reasonably good balance for a launch system that still has some degree of infant mortality.

        • Hi John,

          The Falcon 9 v1.1 was first used on the sixth flight of the “Falcon 9” so its launch record to date is 13/14 (93%), but that is far to small a sample to do a real statistical analysis.

          Your reference to a “fuzzy boundary between cost cutting and quality control” is well taken. Given the fact that SpaceX appears (by what Musk has said) to have accepted the component they say caused the incident without acceptance testing or proof of certification testing would indicate that they have erred far to the “cost cutting” side.

          The “indispensable man” comment is also interesting. Musk is listed (on the press release for the recent telecom) as the SpaceX Founder, Chief (something) Officer and Lead Designer (something to that effect).

          Skipping whether Musk is qualified to be a lead designer on what is essentially a hardware project, to hold that position (make good technical decisions) and be business head of the project (hold cost down) would seem to be a conflict of interest. In an ideal situation those two positions would be held by two well intentioned individuals in constant (friendly) conflict with each other, but not by the same person.

    • “They should have expected better from their supplier.”

      I believe that’s the problem right there – they did expect better from the supplier, and they didn’t have a non-destructive test in place for 100% of parts either in-house or at the supplier capable of spotting the flaw (assuming that’s actually is the issue).

5 Pings & Trackbacks

  1. Pingback:

  2. Pingback:

  3. Pingback:

  4. Pingback:

  5. Pingback:

Sleeping On Alien Shores: The Unquiet Slumbers of the Apollo Moonwalkers

All-Civilian Soyuz TMA-17M Crew Ready for Wednesday Launch to Space Station (Part 2)