38 thoughts on “SpaceX Update”

  1. I’m pretty sure it should be counted as a failed launch, considering that the payload flew ballistically from the top of the stack to the bottom of the stack. However, it’s not a failed launch attempt because they didn’t intend for it to make that trip.

      1. lol. They say Dragon abort would have saved crew but it depends on when the abort was initiated, don’t you think?

        1. Well, if they hadn’t thought about order-of-millisecond LOV events on the pad before, they are certainly thinking about them now!

  2. The only record that matters, to the only people that matter, is this: Was the customer’s payload delivered to an operationally useful orbit (success), mislaid or destroyed (fail), or returned intact with apologies (no test)?

    1. In many of the previous Falcon 9 static tests, the payload wasn’t integrated with the launch vehicle. Looks like this will be a lesson learned.

      1. I believe it’s several decades since a rocket last exploded on the pad. So the lesson would seem to be: don’t make rockets explode.

        As I understand it, testing with the payload in place saves a day or so in the launch schedule, and allows them to verify that the payload will behave as expected when it’s launched. The former may prove to be a big deal if they want to increase their launch rate.

        1. “I believe it’s several decades since a rocket last exploded on the pad.”

          Same thing happened to a manned Soyuz in the 1980’s. The cosmonauts were saved by their launch escape system.

        2. An extra day or two per launch isn’t going to matter much when the company hasn’t been close to maximum launch tempo anyway. SpaceX has demonstrated two-week pad turnaround at least three times at SLC-40. That suggests it will be able to do likewise or pretty close at LC-39A once it’s up and running.

          A minimum turnaround time of 15 or 16 days vs. 14 days only begins to pinch when one gets above 20 annual missions from a single pad. Even assuming a quick find and fix for 9/1’s problem, SpaceX will be fortunate to achieve merely an entry-level double-digit mission total for 2017. LC-39A can handle that. Once SLC-40 is repaired, SpaceX will have two pads in Florida again and 15 or 16-day turnarounds won’t start to pinch until the overall Florida annual mission total hits 40 or so.

          In terms of SpaceX’s overall company launch total, Vandenberg will be a far bigger factor in future than it’s been thus far. Iridium still needs seven deployment missions by the end of 2017. Shotwell, at the recent smallsat conference, took the wraps off a multi-launch deal with smallsat aggregator Spaceflight, Inc.

          Many smallsat constellations are coming and most of the birds need to be in sun-sync or other high-inclination orbits. With their underutilized launch complex at Vandy, SpaceX should be able to successfully land a lot of West Coast gang-deployment business from constellation operators as well as smallsat aggregators.

          Then, in 2018 or 2019, Bolsa Chica comes on-line and enables another dozen missions a year. As early as 2020, SpaceX might well be able to handle an annual mission total equal to roughly its entire current launch manifest.

    2. Exactly. There is a ‘partial success’ category when you did get your payload to orbit, but had to burn the payload’s own reserves to do that, i.e. performance shortfall, insertion accuracy problems. The payload will still pump out revenue for the customer, but less than designed.

  3. It’s curious to note that both the shuttle failures (SRB o-rings and ET) and the Falcon 9 failures (Upperstage helium bottle and upperstage propellant loading) involved the expendable part of the launch stack, not the reusable parts.

  4. The interesting thing to me is future potential. They were already increasing flight rates with one pad. Next year they will have four.

    If stock were publicly available this would have been the time to crack open the piggy bank and buy as much as you could possibly afford.

    These problems are expected. One was quality control from a supplier. This one occurred during fueling which is historically very common even in aircraft (particularly Germany in WW2.)

    You don’t have to be an apologist for SpaceX to be very positive about their future, but the usual suspects will certainly try to use this against them.

    1. Bolsa Chica won’t come on-line until at least 2018. The event of 9/1 probably pushes that back to some time in 2019. But SpaceX still has its Vandy pad and, per their recent statement, LC-39A will be ready in Nov. That’s two. SLC will likely be repaired by some time in 2017. So, one pad now, two by the end of this year, three by the end of next year and four by, say mid-2019. Pad availability doesn’t look like being a real limiting factor barring a 2nd 9/1-style mishap.

      As for finding the cause of 9/1’s explosion, SpaceX is an organizational bulldozer. It will do what it has always done when faced with an unanticipated pile of debris – lower the blade, shift to a lower gear and plow through the obstacle ’til it comes out the other side.

      So I agree SpaceX’s future is bright, however cloudy things seem at the moment.

  5. Does anybody know how the accounting is handled? Payload insurance seems straight forward, but what happens with down payments and accounts payable? Is there any non refundable?

  6. From a damage assessment I’ve read from an eyewitness at the scene the damage to the infrastructure at SLC-40 was severe. It may be a long time if ever if we see launches from there again. Expect the focus to shift to LC39A. I did anyway. If SpaceX does decide to keep ops going at SLC40 I expect it to undergo a major overhaul in order to better address future needs and launch rate support. Now is the perfect time for that. More hangar space, possibly multiple rail access to the pad, hardening or more underground use of important piping & infrastructure. Another key factor and a prime opportunity for SpaceX is the ability to align design and ops with Boca Chica. Making the two launch facilities more uniform in nature increases flexibility, reduces cost, and provides a better degree of redundancy. This actually could be a blessing in disguise.

    1. Calling it a blessing may be going a bit overboard, but I agree it could be an unlooked-for opportunity to make a big batch of lemonade.

    1. SpaceX flies drones as observers normally, no? At least, there’s video from Grasshopper flights and the landing attempts.

    2. Given that that site also claims footage of flying saucers bombing ISIS, I’m a tad dubious. 🙂

      There does appear to be something in the photos taken from that vid, but it appears to be debris from the initial bang.

      I’m not quite willing to rule out intentional sabotage (though I think it highly unlikely). If indeed it’s that, my guess is this was done by a rifle, not a drone. Motive? Lone nut, terrorists, Russians upset that the US was about to overtake them in most launches per year, take your pick.

      My prime suspect remains one of the helium COPV vessels in the second stage.

      1. I still think the initial event looks a lot like a flammable vapor mix just outside the rocket going kaboom. But that raises the question of what the fuel vapor was.

        1. My problem with the flamable vapor theory (I think it possible, I just have issues with it) is the initial event was very powerful, and very bright flash. A fuel-air explosion can do that, but, that’d require aerosolized fuel of some sort. If so, where did it come from? The RP1 and Lox umbilicals are possible, but, they are not where we see the flash; they’re at the base of stage 2. The umbilicals at the level of the flash are electrical (and maybe air conditioning), for the payload.

          The other problem I have with the fuel vapor theory is that was the upwind side of the rocket, so where would enough vapor gather for that big a kaboom?

          Whatever the initial event was, it was very energetic. One of my original suspects was the flight termination system; a static-caused triggering of a detonator which triggered the det cord designed to unzip the tanks. However, the FTS wasn’t armed at the time (they do arm it for static fire, but not that early in the count), and I’ve since been told that the type of arming device they use physically separates the detonators from the det cord when not armed.

          What I’m left with that could provide that much initial energy at that location (Either inside or outside the second stage at that height) is a second stage COPV helium tank letting go in the LOX tank, rupturing the common bulkhead to the RP1. I’m probably wrong, but that’s my guess at the moment.

          1. That’s what I’m thinking, too. I’m sure I’m not the first to suggest this, but maybe SpaceX’s conclusion that CRS-7 was destroyed by a faulty strut was wrong. Perhaps a COPV failed on that mission in such a way that the shrapnel punctured the top bulkhead, releasing the LOX and blowing the Dragon off, but this time the pieces punctured the common bulkhead and the side wall. I don’t know what the ignition source would be, but it wouldn’t take much once the LOX and RP-1 started mixing.

            At least this time they have wreckage to study in addition to examining telemetry. I hope the COPVs don’t turn out to be the culprit. Redesigning struts is one thing, but redesigning the whole pressurization system would be a much bigger headache.

            Also, if it is a COPV, that has unpleasant implications for reusability.

          2. Interesting info about the FTS. FTS was one of my early suspects too.

            One cannot rule anything out at this point, but one potential argument in favor of the COPV hypothesis is that the overpressure event that killed CRS-7 didn’t result in an explosion, just a vapor cloud. That fact, along with the video of the ascent and the initial survival of the Dragon payload, all suggest the O2 tank rupture happened at the top of the tank, not the bottom.

            In that incident, the COPV tank didn’t explode, it tore loose due to a deficient mounting strut and floated to the top of the tank. Its pressurized helium contents didn’t all escape into the O2 tank at once as they were flowing out through torn plumbing.

            The COPV’s are mounted near the bottom of the O2 tank. If one actually ruptured, its contents would all have been relased in a split second and there would be something approximating a blast wave in the O2 tank. Perhaps that, occurring close enough to the O2-RP-1 tank bulkhead, could have ruptured it and allowed a significant propellant mix to both occur and be heated to flash point by the same concussion wave that ruptured the bulkhead.

            The fact that the nearest ullage volume to such a notional rupture would be at the top of the RP-1 tank lends even more credence to such a scenario. A compressible ullage volume on the other side of an incompressible O2 volume through which a concussion wave is travelling would make the intervening bulkhead a weak spot.

            The CRS-7 strut failure was fingered by telemetry data from on-board accelerometers. From a total-energy-initially-released standpoint, CRS-7’s strut breakage was trivial compared to what a COPV rupture would do. I would think such a signature would be about as obvious as bloody shoeprints at a crime scene even quite early in the investigation. If we don’t hear quite soon, even preliminarily, that the root cause was something along these lines, then it probably wasn’t and the search for what did happen might take awhile.

    1. Title notwithstanding, I’m pretty sure that’s 1/25th or 1/30th speed, not 1/250th. But even so, very fast indeed.

      Or at least the visible portion happened very fast; there may have been a somewhat slower internal explosion in the second stage that burst the tank(s) along their entire length between frames. I’ve seen even high-speed video fail to capture tank burst because once the pressure exceeds the failure point a seam can unzip at ~km/s speeds.

    2. huh, Taking it frame by frame like that, the initial flash of the explosion looks more like reflection from a small point light source so bright the video imager was saturated. But the extent of illumination still suggests an explosion outside the rocket itself.

    3. It looks almost like reflected light of a bright explosion, but that can’t be it because the rocket is a cylinder and the light would have to have been reflecting an explosion that was on the tower or someplace else.

      So perhaps there really was a highly flammable mixture pouring down the side of the rocket.

      I don’t think its showing an initial tank rupture because tank ruptures visibly and violently vent fuel or oxidizer before the big boom where those mix.

  9. “Also, as I just noted at Twitter, despite what Wikipedia says, yesterday’s event doesn’t count against their flight record. They’re still 27 of 28.”

    Hyperbole.

      1. I agree with Rand that these are different things. In my world (primarily military aircraft), this kind of loss would go against the mission availability number – i.e., the system is not available to conduct the mission because something broke. Flight record in this context seems to map closely to what I know as mission reliability, which is the probability that the system completes the primary mission objectives once the mission is underway.

        Does the commercial launch vehicle world typically publish the equivalent of a mission availability number?

      2. It is officially a launch failure, because it was an FAA licensed commercial launch, and FAA defines “launch” of an ELV as arrival of the first flight hardware at the launch site. That definition, wacky as it may sound, was not FAA’s ideas. It was the brainchild of the commercial launch companies of the late 1980s, early 1990s. They wanted to have indemnification apply to preflight operations, in case something like this happened and the preflight insurance was insufficient.

  10. Meanwhile back in reality, SpaceX blew up the customer’s payload, destroyed its only operational launch pad, ended its launch operations for the year, and blew a lot of its credibility. The consequences could hardly be more serious. Does anyone think current and potential customers are looking at this and thinking, well after all it’s only a test failure?

    My hopes for the future of HSF are pinned on Musk and SpaceX, but there’s no point in ignoring the obvious.

    1. That’s not the point. Of course customers should be concerned until the root cause is found and addressed. And as for FAA’s loopy definition of “launch” it doesn’t change the reality that it was not, in fact, a launch attempt.

    1. Saw something in that video I messed before, something solid was thrown upward by the initial explosion and can be briefly seen tumbling above the expanding fireball. This object appears to catch fire before the fireball reaches it.

Comments are closed.