SpaceX launch of spy satellite scrubbed

Please consider donating to Behind the Black, by giving either a one-time contribution or a regular subscription, as outlined in the tip jar to the right. Your support will allow me to continue covering science and culture as I have for the past twenty years, independent and free from any outside influence.

The launch by SpaceX of a National Reconnaissance Office spy satellite was scrubbed this morning less than a minute before launch because of an unnamed issue with a sensor in the Falcon 9’s first stage.

The launch has been rescheduled for tomorrow morning, 7 am Eastern.



  • LocalFluff

    Interesting that it only has momentary launch window (or was the sensor warning too complex to fix within hours?). That’s what missions that will rendezvous with a certain spacecraft in orbit, and Lunar missions, have. A spy satellite should be able to launch into its LEO or sunsynchronous orbit anytime. But since it’s launched from KSC, and thus not into polar orbit, maybe it isn’t a traditional spy satellite?

    Meanwhile I learn that the construction of the core module of the permanent Chinese space station is well under way and that it is scheduled to launch early 2019. Maybe ESA’s talk about cooperating with the Chinese on human spaceflight is actually going to happen soon, with European modules and astronauts joining the Chinese. I don’t know of a single mishap in the Chinese human and Lunar programs, other than the Yutu rover halting early. Impressive!

  • Dick Eagleson

    The launch window for NROL-76 is two hours. There are some types of glitches that can be addressed by, say, reloading software, recalibration and/or back-hitching the launch count by an amount that doesn’t bust the launch window. These types of glitches are massively outnumbered by those that require hands-on attention to clear.

    The quick scrub decision is probably checklisted, as should be questionable telemetry from any other component that might need inspection/repair/replacement. Anything, in short, that requires de-tanking the vehicle and a return to the Horizontal Integration Facility for attention. SpaceX has had a number of these overnight glitch-fix episodes. It’s unfortunate, but unlikely to be any bigger a deal this time than in the past.

    And such problems are hardly the exclusive property of SpaceX. ULA recently got triple-whammied by a series of hydraulic issues that, all told, took a solid month to fix. Space is not only hard, it is also, below some level, pretty damned random.

    As for the Europeans anent the Chinese – hoo boy, where to start?

    I guess we could start by trying to nail down a working definition of “soon.”

    Anything involving Europeans on Tiangong 3 is pretty much impossible for a minimum of five years. The Chinese build schedule is pretty slow, even by ISS standards. The first major station piece is scheduled to fly next year, to be followed, roughly annually, by three or four others including a nearby free-flyer Hubble-class telescope.

    Over that same interval, four crew missions and four cargo resupply missions are also scheduled – also annually. Each crew will consist of three people. The time each successive crew spends on-station is set to increase from two months to six over this four year span. That means Tiangong 3 will be uninhabited well over half the time during its first five years. When, if ever, Tiangong 3 becomes continuously inhabited is unclear at best.

    So if ISS can be likened to a small barracks/lab in space, Tiangong 3 looks more like a summer rental.

    Given that there are obvious additional bragging rights to be had by becoming more ISS-like ASAP, I have to assume that the scheduled Chinese effort represents a logistical and financial best effort on their part. To accommodate even one European in addition to the dozen Chinese taikonauts scheduled to serve on Tiangong 3 between now and 2022, would cost China some real money unless said European was a substitute for a taikonaut who would otherwise have gone. I just don’t see that as very realistic.

    Even if the Europeans picked up the entire tab for, say, an extra crew launch and an extra resupply launch, it is by no means clear that the Chinese space industrial base is capable of churning out the needed extra spacecraft.

    And it would be up to the Chinese to shoulder that particular load. The Europeans have no manned spacecraft at present. Their nearest term hypothetical route to acquiring one would be a human-capable Dream Chaser. Making that happen by 2022 would be a stretch even if the go-ahead could be given today.

    Pretty much the same thing applies to any putative European modules for Tiangong 3. There are no such modules extant and ginning any up from scratch would take at least a decade given that ESA is, if anything, even slower than NASA at developing major hardware systems.

    Meanwhile, ISS is, it seems increasingly likely, going to be extended until at least 2028. ESA and Canadian astronauts already get orbital tours of up to six months at a time there. In future, they’re likely to get even more. The Russians are cutting their own ISS staff from three to two. The Russians say this is temporary, but I suspect it will prove to be the new Russian normal. Once Dragon 2 and Starliner are in regular service, the total ISS crew complement can be raised from six to seven. Even if the Russians can keep their former third slot filled with paying space tourists, it seems likely that ESA/Canadian ISS occupancy will rise. It could even get to the point of being essentially a heel-and-toe exercise with a new ESA/Canadian arriving as the previous one prepares to leave.

    That being the case, it’s hard to see why the Europeans would want to subject themselves to considerable additional expense just to be junior partners on a second space station that is only sporadically occupied even by its owners and which would, at best, provide only a fraction of the on-orbit time for their astronauts as would the 2019-and-going-forward ISS.

  • LocalFluff

    Do you have any idea of why a reconnaissance satellite would NOT be launched into a polar orbit?

    ESA talks alot about living in a Moon Village together with the Chinese. It would be prestigious for China to be accepted and normalized and have Europe as a hiker in a Chinese spacecraft dressed in Chinese spacesuits. As you say, the Chinese space station will have empty time. There’s capacity if ESA wants to pay for it. Building space station modules is something ESA has experience from. China seems to have the world’s most reliable space program with clear goals and priorities. What ESA will do is quite random.

  • C. Cecil

    I wonder what the cost is to remove the fuel and oxidizer after a delay like this. Do you suppose the voided fuel tanks remain purged with helium? Any volunteers to change a sensor while the tanks are fully loaded? Just a thought. The pyro-igniters on the space shuttle SRBs were installed on a volunteer basis.

  • Dick Eagleson

    Localfluff first,

    Most imaging reconnaissance satellites are launched into 2-hour sun-synchronous orbits so that the lighting of imaged sites on the ground repeats almost exactly every 24 hours. This makes it easy to spot significant day-to-day changes algorithmically. Sun-synchronous orbits are inclined about 98 degrees from the equatorial plane of Earth, about 8 degrees retrograde. Such orbits also use the differential in Earth’s radius between polar and equatorial regions – and the resulting gravity differential – to precess by a bit less than a degree per day so that the consistent sun angle on the ground at any given site and time of day is maintained as the Earth orbits the Sun.

    Other kinds of surveillance satellites use other orbits. Radar reconnaissance satellites are usually in high-inclination orbits, some being polar. Radar ocean reconnaissance satellites (RORSAT’s) are usually in high-inclination orbits as the polar caps have no surface ship traffic.

    You may recall that one such Soviet satellite fell out of orbit and crashed in the wilderness of Northern Canada in the late 70’s. It was a big deal because the satellite used a 100kW nuclear reactor for power and scattered radioactive material over a wide area near Great Slave Lake.

    Missile launch warning satellites can be in polar orbits because the north polar cap can be breached by nuclear submarines to fire missiles, but other orbits are also used.

    Electronic intelligence (ELINT) satellites are often in geostationary orbit, but some are also in other orbits with high inclinations and much lower altitudes in order to capture signals at high latitudes and from low-power devices such as cell phones and military field radios as well as much higher-powered devices such as air and missile defense radars.

    So-called Molniya orbits – named after the series of Soviet/Russian communications satellites that pioneered this type of orbit, have high apogees on one side of Earth and a low perigee on the other. This makes the satellite “loiter” over the part of Earth beneath its apogee for a large percentage of any given orbit.

    Significant parts of the Soviet Union were at latitudes too high to be served by geostationary communications satellites. Molniyas were designed to augment geostationary birds for military commo. Many types of birds with Molniya-type orbits have since been used by other countries, including the U.S., for both military and civilian purposes.

    One such purpose, as noted widely in coverage of the NROL-76 mission, is as a commo relay between ground sites or low-orbiting recon satellites at high latitudes and geostationary birds that can “see” other ground installations at lower latitudes. This allows real-time commo to and from command and control and intelligence analysis centers to anywhere else on Earth in two hops.

    C. Cecil.

    I doubt it costs much at all to drain the tanks. The propellants are just pumped back into the ground storage tanks they were held in before propellant loading.

    There’s no need to purge a LOX tank with anything. RP-1 tanks can be safed without purging by filling them with nitrogen. Helium is expensive. Nitrogen is cheap.

    SpaceX needs no volunteers to change anything while the tanks are full because, not being idiots – or Russians – they don’t put unprotected human workers anywhere near rockets that are fueled.

    I have no knowledge of Space Shuttle launch procedures, but I’m inclined to doubt that story about the SRB igniters. One can find a lot of stories floating around the Internet about all sorts of things. Only a subset are true.

  • Dick Eagleson wrote, “SpaceX needs no volunteers to change anything while the tanks are full because, not being idiots – or Russians – they don’t put unprotected human workers anywhere near rockets that are fueled.”

    However, SpaceX will do exactly this, with astronauts, when it launches its manned Dragon capsule. But then, so has every other manned mission. You can’t put people in orbit in a rocket that isn’t fueled. :)

  • LocalFluff

    Dick Eagleson,
    Interesting, but I still don’t get what (near) equatorial orbit with a two hours launch window could be useful for a spy satellite.

    Sun synchronous orbit is good for civilian imaging of the ground when the shadows fall the same way each time. But for military uses it seems worthless to observe the area of interest only once every 24 hours. The enemy will simply reschedule their lunch and no activity will ever be seen. Also, Sun synchronous orbits are launched from Vandenberg, not from KSC, right?

    What could cause a two hour launch window? Launches to LEO, SSO, GEO can be done at any time since they do not change relative to the launch site. Maybe because it is part of a constellation of satellites that is not synchronous relative to Earth, but to each other? Like the GPS satellites. Still, two hours is looong, 1½ orbit around Earth in LEO.

  • Mike Borgelt

    Robert, the key word is “unprotected. SpaceX intend to put the people in the rocket THEN fuel it. The astronauts will be able to use the LAS at any time during fueling and after.
    This is NOT what has been done up to now when the rocket has been fueled and then the crew put on board which means both flight crew and assisting ground personnel ARE unprotected human workers near a fueled rocket. Somehow, Tom Stafford thinks this is safer.

  • LocalFluff

    What a beautiful launch today! Because the payload was classified they only followed the first booster online. First stage separation, turn around, boost back burn and puffs from the nitrogen thrusters were all filmed from the ground (or drones) without interruption. It’s becoming a better show each time (in daylight and good weather). Two weeks until next F9.

  • wayne

    Yes– excellent video of the launch and with 1st stage telemetry on the way down.

    Highly informative, appreciate the info.

  • Dick Eagleson


    It occurs to me, upon review, that I didn’t address your points about China and Europe in your second comment.

    You are correct that ESA – or at least its current chief, the man who conceived the Moon Village idea even before he was ESA boss – has made welcoming noises anent Chinese participation in Moon Village. ESA, I suspect, wants China to take a non-trivial part in paying for Moon Village. Given that the idea of Moon Village is completely unaffordable to ESA as a stand-alone entity, the attraction of this idea is understandable. Without a lot of additional money from somewhere besides Europe, Moon Village isn’t ever going to happen.

    As for other potential Euro-Chinese interaction, the Chinese seem much less interested in cooperation with Europe than the Europeans seem to be in cooperation with China. China will do what it plans to do with or without any participation by any other entity.

    It’s also true that ESA’s contractor base has experience building space station stuff, but for ISS, not Tiangong 3. That would be a from-scratch project. Given that ESA is at least as slow and bureaucratic in its internal operations as is NASA, the Tiangong 3 would probably be a decade old by the time ESA could build anything for it.

    Not sure why you regard the Chinese space program as the most reliable. They’ve had no high-profile failures in their manned program thus far, but they also haven’t done much. The unmanned part of the Chinese space program has certainly not been failure-free.

    The Chinese program is, as you correctly note, much more coherent than that of ESA, but is hardly an example of hustle. The currently planned date for a first manned Moon landing, for example, is 2036.

    As for your third comment, NROL-76 isn’t going to a near-equatorial orbit. Based on the launch azimuth, it’s orbit will likely have quite a high inclination – at least 50 degrees – and probably significantly more. We’ll have to wait for the amateur sat spotters to tell us its orbital elements as NRO certainly isn’t going to do so.

    Yes, U.S. sun-sync orbit missions are usually launched from Vandy, but NROL-76 won’t be going into a sun-sync orbit. As for the general usefulness of sun-sync orbits for military recon, you are correct that one such sat can be hidden from fairly easily. But the U.S. doesn’t operate just one sun-sync spysat. It’s a lot more difficult to hide activity from overhead surveillance when there are two, three, four or more sun-sync sats to be ducked.

    I honestly don’t know exactly what considerations went into the establishment of the launch window for NROL-76 at two hours. For that matter, I also don’t know why SpaceX elected to try launching 15 minutes into the window on Sunday and then actually did launch 15 minutes into the widow today. Previous SpaceX missions that had non-instantaneous launch windows always seemed to target liftoff for the very beginning of the window. I suppose I could ask SpaceX or NRO about this, and they might even tell me, but then they’d have to shoot me. I’m content to remain ignorant on this point.

    As to your 4th comment, I agree entirely. The split-screen with tracking camera video on the left and downward-looking 1st stage rocketcam video on the right was a splendid addition. Those nitrogen thrusters are super-impressive from a distance. I had been hoping to see the boostback burn plume impinging on the plume from the 2nd stage from the usual downward-looking engine-cam on the 2nd stage as we saw on CRS-10. But the side view of the impinging plumes from the tracking cam was at least as good. SpaceX does know how to put on a show.

    Mr. Zimmerman,

    I don’t think I can improve on what Mike Borgelt wrote. I view SpaceX’s approach to human spaceflight operations as being fundamentally superior, from a safety standpoint, to anything NASA has ever done, and for the reasons Mike laid out. Good on ya, Mike, as the Aussies like to say.


    Thanks for the kind words.

  • Edward

    C. Cecil wrote: “The pyro-igniters on the space shuttle SRBs were installed on a volunteer basis.”

    My understanding of the SRB igniters is that they were built into the top of the SRB stack. Igniter installation would be during the stack build-up, not during pad operations.

    Although solid rockets are seen as relatively safe (people mount them as JATO rockets onto aircraft), they can go terribly wrong. In 2003 the Brazilians had a solid rocket accident while a couple dozen people were working on a rocket, killing most or all of them.

    During the Space Shuttle days, NASA did not give tours of the VAB, because there were SRB sections stored in the building, and NASA did not consider them to be safe enough to let tourists inside the building at the same time that they were there. The Brazilian incident makes clear that the decision was correct.

    The Russian Nedelin explosion also informs us that we must limit the number of people near a fueled rocket.

    The extra chilled liquid propellant that SpaceX uses on Falcon 9 needs to be loaded at the last minute, in order for it to remain cold and dense enough for some of SpaceX’s launches. This may also limit the launch window for Falcon 9 and Falcon Heavy. This could also give the ground crew for manned missions time to load the astronauts and leave the pad before propellant loading commences. As Mike Borgelt wrote, the astronauts would be relatively safe, as their launch escape system can quickly take them clear of any pad explosion.

    You wrote: “But for military uses it seems worthless to observe the area of interest only once every 24 hours. The enemy will simply reschedule their lunch and no activity will ever be seen.

    Any satellite’s orbit can be predicted and ground activity stopped in order to prevent being seen. The advantage is that overall activity (e.g. construction) can be viewed day after day. Also, The US has satellites in geostationary orbit that watch for enemy rocket launches, so all launch activity will be seen.

    LocalFluff wrote: “What could cause a two hour launch window?

    Several things contribute to the length of a launch window. I once worked on a satellite that had a one-second launch window. If it didn’t launch on time, then the launch would have to be delayed to the next day.

Leave a Reply

Your email address will not be published. Required fields are marked *