40 thoughts on “Tomorrow’s Launch”

  2. They always talk about the weather for launch, never the weather for the landing on the drone ship. But the recent picture showing it in position seems to show very mild weather at sea.

  3. I’m watching it from the comfort of my bedroom on the webcast, admiring how thoroughly it’s fogged in.

  4. Another crash on deck. Broken landing gear. Call Grumman and get a new gear designed for Carrier Landings. The current design always seemed too delicate for Naval use.

    If you drop an empty 1st stage on a floating barge the barge will move. Then it will roll back and kick you in the ass. What is the landing weight of the stage at set down? I’ll bet its the heaviest thing ever landed on a floating vessel.

    BD, Former Naval Flight Officer
    Lt., VP-47, 1978-1981

    1. You betcha Bill. World of difference between several thousand feet of concrete and a pitching deck. Seas were running 12-15 at the barge. I’m confident SpaceX will get this right eventually and when they do they will be that much further ahead of their competitors.

      1. It’s big, but is it just a barge or a semi-sub. I think they will eventually need to go semi-sub to negate wave action. Cheaper to do that than making stronger gear in the long run. I bet Elon could get a good deal on a floating oil rig right about now.

        1. I should note… according to Elon, this failure wasn’t from lack of structural capability, but failure to fully lock the gear down. In other words, impact velocity was within tolerance if gear was nominal.

          However, I still think that long term operations would be better supported by semi-sub (which could double as a launch facility… and I know, Boeing tried it but they disposed of the rockets).

    3. The dry mass for a F9 1.1 is about 25 tons, and at landing it’s pretty much at dry mass. For comparison, a F-18 is about 14 tons. But, the record for heaviest carrier landing is IMHO probably the 1963 landing of a C-130, at 70 tons. (Yep, they landed a Herc on USS Forrestal).

      Per a tweet from Elon Musk, one of the landing legs didn’t lock. What I don’t know is how different the landing legs are on the F9 1.1 to the F9 Full Thrust (the latter did the RTLS). If the problem turns out to be inherent to the 1.1, it’s not a problem because Jason 3 was the last of them.

      Hrmmm, perhaps the problem here is spin. Instead of looking at this as a failed landing, perhaps we should hail the F9 as the most effective anti-ASDS (droneship) weapons system ever made, with a 3 for 3 success ratio. 🙂

      1. Yep, they landed a Herc on USS Forrestal

        Those were awesome trials! There are a view videos of them them on YouTube!

    4. The Space Launch Report puts the total mass of the First stage at about 440 tons. With Musk having said the mass ratio of the First stage is 30/1, the dry mass alone must be under 15 tons. Add about 5 seconds more propellant for the single engine firing at 70 percent thrust and an isp of 282, that’s just .162 tons added to dry mass.

      I remember the Navy’s Vigilante supersonic Bomber back in the late 1950s having a dry mass way more than that, much less more modern warbirds. So, no, it’s not the mass. IMHO, its probably the movement of the barge.

    5. I’m with you on this one BD. I’ve heard carrier landings should be more accurately called controlled crashes. You can’t gun the engines on an F9 in case of hook failure so the best option is more flexible legs. You don’t parachute onto stiff legs for similar reasons.

  5. Comment from Musk is that a latch on one of the legs didn’t engage.

    This is one of those unknowns I was talking about.

    1. Good to know that its a correctable error, then. If they did stick the landing but for that, then barges are not impossible!

    2. As they knock down these unknowns one after another – a corroded nut here, a broken strut there, a malfunctioning latch now – the list of unknowns will shrink. They are really close to nailing this.

      1. Corroded nuts, and broken struts are not unknowns. They are bad operational technique.

        Unknowns CAN be eliminated once they are knowns.

        An example unknown would be, say something like an engine vibration causes something to fail. And they had to track down the vibration as the cause.
        Once tracked down they can redesign the failed part or eliminate the vibes.

        A corroded nut would be something that was missed.

        1. Thought now was that it was ice buildup due to the heavy fog, and this intereferred with the latch.

          Unknowns can be eliminated, but this has a cost. The marginal cost of the testing they are doing is probably low, since they would expend the stages anyway. If testing is cheap then it makes sense to test rather than try to bury all potential problems in mountains of reports.

          1. Many failures have occurred by not testing in the environment of Vandenberg. The first (and last) Amroc launch attempt failed as a result of assuming that the LOX valve that had been tested many times in the Mojave desert wouldn’t freeze open at the coast.

          2. I guess the question I have, after having watched the (almost) landing video, is how similar are the landing systems between this older Falcon 9 and the newer? Are the latching mechanisms similar, such that this is truly a learning experience or is it merely a one-off? Musk hasn’t given us enough information to know. Also the ice speculation could be relevant but could there have been other factors that weakened the latching? Such a a heave or deck roll that increased forces on this strut above the others? Another question in my mind is whether enough of the booster remains after the explosion to tell? Or whether key components got ejected overboard? Well this is all ill-informed speculation on my part. It would be interesting to know how well instrumented the ASDS is in regards to deck forces being experienced at the moment of landing. There was some mention in the past of using islands off the west coast of California that are operated by the Navy as potential land-based Landing Zones as well. There is no doubt about the complexity of at sea landings.

          3. The thing that had me scratching my head this morning is how ice would survived launch, ascent and entry. It would have had to be formed during the landing itself. Maybe just install some heaters?

          4. The thing that had me scratching my head this morning is how ice would survived launch, ascent and entry.

            I remember a flight on a 737 many years ago. There was some green fluid* on the wing in a stagnant area. We were doing over 500 MPH and yet the fluid stayed in place. If it was ice, it might’ve been in an area of stagnant airflow where the heating of launch and reentry couldn’t reach it.

            *I notified a flight attendant and she told the flight crew. The copilot came and checked the situation. I don’t remember if green fluid is normally turbine engine oil or hydraulic fluid. We flew on to our destination and the flight landed uneventfully.

      1. I agree that it would not have helped, if the only thing wrong was that the latch didn’t work. Even if barge motion contributed to that, Solving the movement of what is a probably less than 1 kilo latch by adding motion under guidance to a 100+tons platform seems, …excessive. Better to change the latch!

        1. Ya, Leland’s idea is better.

          Seems to me not having to worry about how rough the seas are or the stress they would put on the rocket would help their overall operations but I’m clearly not an engineer.

          1. It is another layer of complexity. That isn’t to say it isn’t a good idea, just that every new layer of complexity adds to the number of things that can go wrong, the number of things that have to be maintained, etc.

            They are going to want to keep things as simple as possible. They have already demonstrated that they can land with accuracy. Instead of four legs, maybe six would be better. Or maybe a big damn net strung up around the platform.

  8. There’s a reason office chairs are moving toward five legs instead of four. Could NASA even man-rate a four legged office chair these days? The risk is just too great.

    1. My FIRST thought on watching the video was “a five-legged configuration would have stayed upright.” Wonder if they did trade studies… useless weight most of the time, but occasionally saves the vehicle?

      1. Hmm – are you sure? Those legs are pretty wide and the Falcon is really tall, and it has to stay upright on a gently (?) tossing drone ship for a hour or whatever till they can get men on board to lock the legs down. You might need eight legs or such.
        I was always amazed that they expected to be able to bring it in to port at all, vertical like that.

  9. The non-reusable Falcon 9 is listed as being able to deliver 13,150 kg to LEO. Jason-3 is listed at 553 kg ($113,000 / kg to LEO). The last Orbcomm was, I believe only about 2,000 kg ($31,250 / kg to LEO). So, does anyone here have a sense as to whether a Falcon 9 reusable would be able to deliver anything close to 13,150 kg to LEO? Do we have evidence that reusable F9s would actually save money? Just askin’.

    2. My performance model, at http://www.silverbirdastronautics.com/LVperform.html , gives 13.5 tonnes to 185 km / 28.5 degrees for a Falcon Full Thrust with first-stage RTLS. There’s more guesswork in there than I’d like, as SpaceX is being particularly stingy with the relevant data, but it should be within 10-20%. Basically, Full Thrust seems to have been designed to recover the performance that would otherwise be lost with recovery.

      Jason, was the only satellite going to that orbit this year, so it paid full price for a mostly-empty Falcon. This should not be taken as an indication of the system’s actual performance.

    3. Doug, SpaceX, and all other launch companies, charge by the launch, not by the Kilo. Price per Kilo to orbit is calculated on the basis of the customer making full use of the payload capability, though. If they don’t load up the rocket fully, they still pay full price for the rocket.

      both Musk and several others at SpaceX have estimated that the drop in payload to LEO would be 30% for both stages being reusable, and about 15% for just the first stage. That does *not* mean the total lifetime of the vehicles, just the single mission payload capability.

      Thus, if, as Musk estimates, the un-reusable first stage is 75% of cost to SpaceX for each launch, and that reusable first stages can be re-used 10 times before thorough refurbishment, then SpaceX will save $45 million out of each launch for those 10 launches. That means the company will save $450million for those 10 launches, while delivering 15% less than 131,500 KG., *if*needed*. Since .85 times 131,500 is 111,775KG. However, SpaceX need only spend the present cost, minus the $450 million for those 10 launches.

      If their business model says they will profit more by handing some of that 75% back to customers, and expanding launches thereby, then they will. Otherwise, it goes straight into the SpaceX bank accounts. If *all* of that payload capability is actually used by their customers, and if SpaceX’s CFO is convinced it will be better for the company, then SpaceX could drop their price by nearly that much. SpaceX can then claim they launched those payloads at a price per kilo equaling only about $150 million divided by 111,775 KG, and get a price per kilo of $1341 per kilo to orbit, or a bit more, since their ground crew costs don’t drop much if any.

      1. I am glad you made this post because I made the stupidest of stupid mistakes in my spreadsheet.

        Assumptions: original price = $70,000,000, launch 1 = $62,100,000, launch 2-10 = $15,525 each, and 11,178kg to LEO.

        Going by your scenario and there are no refurb costs for those 10 launches but that SpaceX wants to make the same amount of profit off each launch as they do now, which is $7.9m, they can charge $23,425,000 per launch for $2,096/kg.

        But if you share the cost of the first stage across 10 launches, they can charge @ $28,082,500 per launch with an average cost of $20,182,500. Taking into account the 15% performance hit, this means its about $2500/kg to LEO for those 10 launches.

        Say they spend $1m on refurb each launch (no idea how realistic that is) and they can fly a core 10 times a year. The average cost for those 10 launches are then $21,082,500 and they can charge $28,982,500 to make the same profit and its $2593/kg to LEO.

        Using those same numbers and assumptions we can compare to the 2016 Orion development budget of $1.2b. SpaceX could do 40 launches of 4 cores @ $28,982,500. That leaves $40m left over which isn’t enough to pay for another launch of a new core but we will give it to them anyway. This means SpaceX could put 447,100kg in LEO for $2684/kg.

        Then if the SLS block I can put 70,000kg in LEO, for what we spend on Orion development in 2016, SpaceX could do the equivalent of 6.39 SLS launches. It might be a while before SpaceX could do that but I haven’t seen anyone claim SLS could do 6 launches in a year, ever. Also, this leaves $2b from SLS that could be spent on payloads. How long until SLS flies? :*(

  10. What do they mean by latch? Is there a latch that secures the leg to the deck or did the leg collapse?

    In the after pictures, there isn’t any waves to speak of but someone on twitter posted a gif fromthe barge POV that showed lots of motion.

Comments are closed.