Longer Video Shows How Incredibly Close Falcon Stage Came To Successful Landing

Bruce Perens writes In the video here, the Falcon 9 first stage is shown landing with a tilt, and then a thruster keeps the rocket vertical on the barge for a few seconds before it quits, followed by Kabooom with obvious significant damage to the barge. It looks like this attempt was incredibly close to success. Given fixes, a successful first-stage recovery seems likely.

Comments

Larger landing area

[Dan East]

It sure seems that if a larger landing area was available, so that the rocket didn't have to lean so far to adjust to a very small target and thus could prioritize staying vertical, it would be able to land successfully. What's it going to take for NASA or the FAA or whatever to give them permission to land on, um, land.

Re:Larger landing area

[Bruce Perens]

The ship is 300 feet long. It's a big rocket :-)

The pad area they have at KSC is made for F9 Heavy, and multiple stages are supposed to land there, the neighbors are sensitive about having other rockets come down in their yard, and there's a big building you really don't want to hit :-) . So, they probably do need the precision. There was an odd tweet from Musk, later deleted, that said there was actually a process control problem and a phase delay.

Re:Larger landing area

[Megane]

I had heard that a valve got stuck, causing the throttle and gymbaling to get out of sync. Now that I see the video, it looks like it was coming down really well until that last moment. It probably wanted to make a final course adjustment, but the rocket bits didn't work the way the computer bits expected them to. It even seemed a bit like the kind of crashes I would get in Gravitar after getting a bit confused.

A larger landing area would have just meant no bits to fall into the water after the kaboom.

Re:Larger landing area

[narf0708]

It sure seems that if a larger landing area was available, so that the rocket didn't have to lean so far to adjust to a very small target and thus could prioritize staying vertical, it would be able to land successfully.

Or they could get their engineers to take a look at the PID system they're using so that the rocket could correct for the error by making more gradual adjustments sooner, it wouldn't have to lean so far.

Re:Larger landing area

[Bruce Perens]

The tweet from Musk this morning used the word "sticktion", meaning static friction. And said it was the cause of a phase delay. And then the tweet got deleted.

Re:Larger landing area

[narf0708]

The pad area they have at KSC...

The Kerbal Space Center?

Re:Larger landing area

[Bruce Perens]

The Kerbal Space Center?

Not until the next time we cut the NASA budget to pay for a subsidy of some incredibly rich industry. Like oil. We need more oil drilling subsidies, don't we? Or intellectual property. That's just another word for innovation, isn't it?

Re:Larger landing area

[farble1670]

Now that I see the video, it looks like it was coming down really well until that last moment.

no way. for a gigantic fragile rocket it came in extremely hot. it was probably moving at 50+ MPH when reached 50 feet of the platform. it didn't slow down much until it was less than 10 feet away. it was also wobbling as it came in.

as much as i'd like to say that was close, it wasn't.

Re:Larger landing area

[Anonymous Coward]

Compared to where the first stage started its landing cycle, the orientation, velocity, and location of this not-quite-a-landing was INCREDIBLY close.

Re:Larger landing area

[towermac]

It's not all politics.

The first stage is a good distance downrange when it is done, as in; way out in the Atlantic when launching from Cape Canaveral.

So, to fly it back to Kennedy would take a lot more fuel than letting it come down where it is. Not to mention the 'just in case' scenarios involving Titusville, the nearby I-95, etc.

It will benefit Elon greatly if he can pull off landing it on his ocean barge. If he can do that, he can launch from anywhere in the world, with competitive fuel/payload ratios, and still reuse his rockets.

But yes, overall, it would be easier to land it on land.

Re:Larger landing area

[Dunbal]

You are assuming unlimited fuel...

Re:Larger landing area

[towermac]

ty for that.

Re:Larger landing area

[Bruce Perens]

That would be hot for an aircraft, but it was the planned vertical speed profile for the rocket. The grid fins need speed to work and they are the main control surfaces. The cold gas thrusters don't have infinite gas behind them and the engine burns are very short.

Re:Larger landing area

[Anonymous Coward]

That large explosion at the end of the longer video implies there was plenty of fuel left over.

Re: Larger landing area

[D.McG.]

The full tweet was as follows:
@elonmusk: @ID_AA_Carmack Looks like the issue was stiction in the biprop throttle valve, resulting in control system phase lag. Should be easy to fix.

Re:Larger landing area

[Bruce Perens]

I was surprised by something in the re-entry profile. They use what they call "lift" from tilting the rocket body against the air stream to control horizontal motion. I call it "falling with style". So they can go back uprange some distance without an additional fuel expenditure.

All of their communication so far has been that they can get back to the pad with the F9 or the two outer stages of the F9 Heavy. The center stage of F9 Heavy would probably need the barge.

Re:Larger landing area

[Anonymous Coward]

My guess is it's intended to be a suicide burn.

Re:Larger landing area

[jo_ham]

That large explosion at the end of the longer video implies there was plenty of fuel left over.

You would assume that, yes. But a) that's mainly the fuel vapour in the near-empty main tanks, and b) the fuel the GP is talking about is the pressurisation of the hydraulic lines for the control systems that run the thrusters and fins, not the fuel that runs the rocket engine.

Re:Larger landing area

[voidptr]

The F9 is intended to land with what they call a "Hover Slam" maneuver - the engines decelerate it to zero right above the surface in as little time as possible. The Merlin engines have a limited throttle range, and with the stage empty, just one engine firing at the lowest throttle setting has a thrust-to-weight ratio somewhere around 1.8, so it can't hover. It would decelerate to zero and then start to lift off again if the engine isn't shut off, you'd need a TWR of 1.0 to just counter gravity and make it hover.

Re:Larger landing area

[Hiroto. S]

That would be hot for an aircraft, but it was the planned vertical speed profile for the rocket.

Wasn't it supposed to land at 5m/s? First stage is about 37m tall and it should take leisurely 5 - 6 seconds or so to finish the last rocket length, but in the video, it is finishing that in about 2.5 - 3 sec, coming in at least twice as too hot. Maybe they need to start the landing thrust several seconds earlier, assuming they have enough fuel left.

Re:Larger landing area

[BoogieChile]

So really all they need is to have a Kerbal aboard to give it a flick with a fingernail to get it started.

Re:Larger landing area

[pushing-robot]

Rockets are capable of incredible acceleration, especially when they're low on fuel and deprived of their payload. Under those conditions, the F9 first stage could easily go from 50MPH (~22m/s) to 0 in the space of a few meters.

Also, you *want* to land fast, because for every second you spend in the air you lose another 10m/s of your limited delta-v (fuel), and the faster you're traveling the more aerodynamic control you have.

Yes, I know all this from playing KSP.

Re:Larger landing area

[Jane Q. Public]

Or pee on it to loosen up that static friction.

Re:Larger landing area

[Applehu Akbar]

There needs to be a capture system at the landing site. This could be something as simple as a slightly conical pit about two-thirds as deep as the stage is long, with "soft" sides. This could include a net that gets thrown over the stage right after touchdown.

Re:Larger landing area

[nedlohs]

That's how you do it. A suicide burn or "hover slam" is the best option otherwise you use too much fuel. You want to burn as late as possible - the ideal would be a full throttle burn as low as possible that puts at 0 velocity at exactly the touchdown point, but of course you won't pull that off.

They also want to be "off target" until very late since if it engine fails you'd rather it hit the water not the landing platform at very high velocity.

Re:Larger landing area

[Anonymous Coward]

Would that be the same FAA that won't even let you fly a little quad-copter for profit?

Re:Larger landing area

[JWSmythe]

I'm thinking they need to figure out a better way rather than landing it vertical. Maybe when they get it that close, they could do some sort of net capture, rather than hoping it will stay upright. It would solve some of the more delicate problems. That could create all kinds of new problems though.

Re:Larger landing area

[JWSmythe]

So you're saying it wasn't hyper quantum sticktation of the flux attitude gimble during the multiphasic delay sequence?

Damn, I should write science fiction. Or maybe I could work for the media. Both string scientific sounding words that mean absolutely nothing. :)

Re:Larger landing area

[gman003]

Horizontal velocity would still need to be killed. If it has too much sideways momentum at touchdown, it'll just drag and tip over, no matter how big the landing area.

SpaceX is, unsurprisingly, doing things the right way. The hard way, to be sure, but when they finally get it working by proper engineering, rather than overkill, it will be well worth it.

Re: Larger landing area

[garyisabusyguy]

Carmack joined Space-X?!

Re:Larger landing area

[Bruce Perens]

It's still slowing down during the last rocket length. That is really cutting it close, yes. I think the goal is to use an absolutely minimal fuel expenditure. The current configuration is not capable of landing after a GTO insertion. When they were considering doing the test for the DISCOVR flight, they were not going to have enough fuel for the normal recovery sequence, and were planning to delete the subsonic decel burn and come up to the barge at 1 KPS rather than the leisurely 250 m/s.

Re:Larger landing area

[0123456]

In the official Falcon Heavy video on Youtube, the stages seem to touch down much slower, so I presume they're planning to modify the engines for deeper throttling in future.

Re: Larger landing area

[He Who Has No Name]

Armadillo Aerospace had their own prototype hovering-gimballing single engine rocket, and Carmack was heavily involved in development of the logic and systems that made that possible.

Re:Larger landing area

[jordanjay29]

As recently proven by Amazon, you can easily twist the FAA's arm if you're a big, multi-national corporation. SpaceX shouldn't have a difficult time of it.

Re:Larger landing area

[He Who Has No Name]

That's exactly what it is, since they don't have the throttle authority to burn lower, longer.

With that approach profile, they're aiming to have dV hit zero the same exact instant that the rocket settles into a 1g static load on its landing gear - same as if it was just sitting on the ground.

There's a good reason it's called a suicide burn.

Re:Larger landing area

[Anonymous Coward]

Same fuel for both. It's just vented out to pressurize the hydraulics.

Re:Larger landing area

[G-forze]

No, the problem is the Merlin D engine produces way too much thrust for the empty first stage to be able to hover. They can only do a really short break at the end or the whole thing will begin to move upwards again.

Re:Larger landing area

[Anonymous Coward]

I think SpaceX is showing what can be done on a shoestring budget with ingenuity - Nasa has done some really cool stuff, but they're a huge bureaucracy that isn't all that efficient.

If you gave NASA's budget to Elon Musk - we'd have humans on mars before 2020

Re:Larger landing area

[Anonymous Coward]

I originally thought it was a matter of the downrange distance as well, but after some reading I discovered they are indeed planning to eventually fly all the way back. This isn't like a Saturn V that's halfway across the Atlantic at first stage separation - the Falcon 9 is only about 100km out and just about as high at staging, and nowhere near orbital velocity. The landing barge they're using now is 300km downrange - they fly on a ballistic trajectory for a bit before restarting the engines, possibly to give them time to analyze telemetrics before deciding if the landing is go. Their current flight path already includes a U-turn at altitude so the rocket is moving back toward the launch pad before re-entering the lower atmosphere and landing.

Re:Larger landing area

[stealth_finger]

Now that I see the video, it looks like it was coming down really well until that last moment.

no way. for a gigantic fragile rocket it came in extremely hot. it was probably moving at 50+ MPH when reached 50 feet of the platform. it didn't slow down much until it was less than 10 feet away. it was also wobbling as it came in.

as much as i'd like to say that was close, it wasn't.

It hit the small boat in a big ocean, almost upright, and almost stabilised itself. I'd say that was pretty damn close.

Re:Larger landing area

[Anonymous Coward]

It will benefit Elon greatly if he can pull off landing it on his ocean barge. If he can do that, he can launch from anywhere in the world, with competitive fuel/payload ratios, and still reuse his rockets.

Why must it land on the barge? Seems like they need to bring extra fuel and complicated guidance systems just for that. The simpler approach would be to rocket-brake only to the point where you can deploy a parachute, and land it in the ocean near the barge. Then the barge/ship picks up the sea-cooled rocket using a crane. You need an extra crane - but it is not part of the rocket so much cheaper to bring around.

Landing directly on the barge means any people will have to be on another boat for safety - so you need another boat. And what if it lands perfectly? Then you have to secure a very hot rocket, before the waves & wind has time to tip it. A few seconds in rough weather.

Re:Larger landing area

[khallow]

Because salt water is bad for rocket engines.

Re:Larger landing area

[Sardaukar86]

Thanks Bruce, it's often in these little nuances that I catch myself staring into space, contemplating the sheer enormity of what has been achieved here. The skills of these people - to do what they're doing with the budget at their disposal - almost completely wrecks my personal 'scale of difficulty'. I thought I understood what they meant by 'It really IS rocket science,' but I'm really not even close.

We really must be living in the future: small, agile, private enterprises taking the reigns of progress from state-level actors. NASA are by no means obsolete, if anything they've adapted rather well for a bureaucracy of their size and are continuing to do amazing science.

Budgets might be tighter than we'd like but I can't help feeling like we're entering a golden era of space exploration and related technologies.

Ooooh.. uh, does this make me a Space Nutter?

Re:Larger landing area

[Altanar]

Some time in the future, SpaceX *does* plan on performing landing at Cape Canaveral. They already have control of what was formerly called Launch Complex 13. SpaceX has now renamed it "Landing Complex 1".

Re:Larger landing area

[trout007]

It will be the AirForce that makes the call since they run the Missle Range at Cape Canaveral. The range is designed to handle these types of accidents. The pads are seperated by miles of scrub brush. They also have flight termination system if the rocket goes outside of the flight path they destroy it.

I'm thinking it's going to be pretty soon they will be able to attempt a landing there.

Re:Larger landing area

[CrimsonAvenger]

Rockets are capable of incredible acceleration, especially when they're low on fuel and deprived of their payload. Under those conditions, the F9 first stage could easily go from 50MPH (~22m/s) to 0 in the space of a few meters.

Specifically, from a starting height of 50 feet, and a starting velocity of 50 mph downward, it would require a net acceleration of ~16.5 m/s^2 to come to a stop at ground level.

Since a single (much less nine) Merlin engine can manage 654 kN thrust at sealevel, and a (nearly) empty Falcon 9 first stage masses under 20000 kg, a Falcon first stage is capable of >32.7 m/s^2 acceleration (assuming only one engine burning, of course). Which is more than plenty to allow it to come to a stop on the ground from the estimated speed/height of the OP....

Re: Larger landing area

[michelcolman]

Certainly looked that way to me. On the last oscillation before touchdown, with the tail end moving towards the left, the thrusters keep pointing the same way as the rocket goes through vertical and only change direction a little bit afterwards. This increased the amplitude of the oscillation rather than decreasing it. The thrusters should have changed direction before passing through vertical, not afterwards. I can't imagine them getting this wrong in software, it's basic dynamic stability 101, so a sticky valve seems likely.

The rocket ended up landing almost perfectly vertical, but still rotating so the base was traveling sideways over the landing pad. No way they could stay upright like that.

Re:Larger landing area

[AmiMoJo]

So can someone explain why they don't use a soft landing pad? Maybe an inflatable. It would have to be robust enough to survive the engine firing near it, but would allow the rocket to land on its side instead of having to stand upright.

Re:Larger landing area

[Flentil]

I think because the rocket was designed to land upright. That's the thing it does that makes it so special.

Re:Larger landing area

[Anonymous Coward]

Kennedy Space Center.

I was confused too, last time we discussed NASA, but someone in that discussion did bother to write it out.

Re:Larger landing area

[meta-monkey]

Also, do main engines have precise thrust control? It seems more like a full-on, full-off kind of deal.

Re: Larger landing area

[whistlepig]

This is a simplified explanation, but also the first one I have seen here. Mod parent up.

Re:Larger landing area

[anegg]

Not necessarily. Too much money can be just as much of a problem as too little. Budget constraints can promote clearer thinking and more clever innovations.

Re:Larger landing area

[Andy Dodd]

I think their minimum is something like 50-60% - and I'm not sure if there are any throttle levels in between "min" and "max"

Re:Larger landing area

[CastrTroy]

If you haven't already, read An Astronaut's Guide To Life On Earth, but Chris Hadfield. Amazing book. It gave me a new found respect for NASA, the CSA, Roscosmos, and everybody else who helps make space exploration possible.

Re:Larger landing area

[tp1024]

They tried parachutes - didin't work. When they did land it in the sea, it tipped over and broke.

They sure would liked to have landed it on land right away, but have a look at what happened on the barge and now think about how likely it is you would get landing permit for an unproven rocket. You see, you won't get it. So the main thing they are doing right now is demonstrating that they can land it accurately and the maneuver is safe for the public.

They also want to land it on the ocean for performance reasons. Flying the rocket back to land takes a lot of fuel you'd rather use to further accelerate the 2nd stage, if you have a heavy payload.

Re:Larger landing area

[Isca]

Think of it as a soda can. When the rocket is pressurized with all of it's fuel it's 100 tons and can withstand very heavy vertical forces. When most of it's fuel is gone it's like an aluminum can - still strong top to bottom but on the side very easy to dent and damage. 90% of the 10 tons of weight left is in the bottom 10% of the rocket so it's not as tipsy as you'd think.

Re:Larger landing area

[DerekLyons]

It sure seems that if a larger landing area was available, so that the rocket didn't have to lean so far to adjust to a very small target and thus could prioritize staying vertical, it would be able to land successfully.

No so much as you might think - you still have to trim and eventually null your horizontal velocity, and null any horizontal residuals arising from trying to remain vertical. It's a complicated problem, even if you're just aiming for an arbitrary landing spot in a larger landing area.

On top of that, the crash seems to have been caused not by prioritization, but by a control valve operating sluggishly causing the response time to go out of limits. Even if you're just accepting a landing wherever you're coming down, if the control system gets out of phase you're screwed.

Re:Larger landing area

[serviscope_minor]

so I presume they're planning to modify the engines for deeper throttling in future.

That could be tricky. The reason it's hard to deep throttle rockets is that in order to work they have to choke. If you look at a rocket, they have a chamber which pinches town to a small diameter then grows again into the bell. It looks like a Venturi, but isn't: it's necessary for a shock wave to form at the narrow point as the combusting gasses go supersonic on the way out.

If the feed into the engine is too low then it will un-choke and resemble a large, exensive BBQ than an actual rocket. At that point the efficiency will go through the floor and the thrust will drop to almost nothing.

Re:Larger landing area

[Bruce Perens]

Yes, but what engine were they flying in F9R and Grasshopper? Both of those were lighted up for the entire flight and didn't bounce around.

Re:Larger landing area

[trybywrench]

Isn't this what one of the Mars Rovers did? They "flew" it to a landing area by just tilting and then controlling the tilt IIRC. Oh and I completely agree with Sardaukar86, whenever I start thinking I'm hot shit I go watch some SpaceX videos and I'm back to my humble self.

Re:Larger landing area

[Bruce Perens]

I learned an important lesson from Open Source, and it applies to SpaceX too: Things work a lot better if you just give the engineers the freedom to do engineering.

I think that in government projects and in most larger companies we tend to devalue the technical people in favor of the nontechnical. And we don't give them much power to actually run things. And then we wonder why efficiency is so low.

It has certainly given me ideas for how to run my own company.

Re:Larger landing area

[phayes]

F9R & grasshopper were ballasted to a weight within the throttle limits of the engines, even empty...

Re:Larger landing area

[Gavagai80]

I doubt anything is soft enough to tip over a 70 meter high rocket and not result in damage while falling.

Re:Larger landing area

[Gavagai80]

Wouldn't the falcon heavy have more control over the acceleration rate when using the same engines, simply because it's heavier with 27 engines instead of 9? To control the rate of descent, just select the number of engines to fire.

Re:Larger landing area

[Coren22]

Is that Kerbal Per Second?

Re:Larger landing area

[recharged95]

I hope that part of landing you describe is not under closed loop. If so, that's why Elon's tweet was removed....

It's typical for any vertical landing mechanism [on Earth] to go open loop. Most quadcopters (much like ours) have enough throttle authority until entering ground effect, rockets in vertical landing also apply, but it's a different type of ground effect as you can see in the video--that thing had a high descent rate likely to keep it from horizontally drifting off target. In our copter's case, once we enter ground effect, our s/w goes into a open loop algo and it does a similar hover-slam technique--why? again, to keep it from drifting horizontally (likely induced from ground effect), aka on target.

Once you enter ground effect, all models will likely fail using a closed loop approach, and you need to go open loop, vertical dead reckon and use gravity to your advantage. If they are expecting oscillations upon touchdown (i.e. Elon's phase lag comment), there's something else wrong, or their approach (again, closed loop?) isn't right: you shouldn't design to have oscillations, especially when it's applied in ground effect.

Re:Larger landing area

[cbhacking]

Irony: the rocket motor with the highest TWR in the industry (at over 150, the Merlin 1D that SpaceX uses beats any of their competition), which is normally a very good thing, has *too much* TWR, even with only one engine out of nine firing!

Re:Larger landing area

[Capt.Albatross]

That's an interesting comment about the phase delay. My impression as I watched the video was that the rocket behavior resembled the pilot-induced oscillations that can occur when a pilot's responses lag too far behind the events.

Re:Larger landing area

[lucien86]

Exactly - great explanation.. What they need is a smaller central landing engine that can give the right amount of throttle. Otherwise I can see them ever getting a good success ratio using this method.

I like the method recently chosen for Vulcan, eject the engine cluster and recover it by parachute and inflatable landing system. Simpler and easier, but it doesn't recover the whole stage..

Re:Larger landing area

[Anonymous Coward]

The oil industry is "learning" this now. That's in quotes because we should've learned it in the last 15 cycles. But damn the torpedoes and borrow borrow borrow turns into "how can we get leaner" when the revenue stream dries up. Does actually drive innovation though.

Re:Larger landing area

[Sardaukar86]

Thanks CastrTroy, I'll pick up a copy - much appreciated!

Re:Larger landing area

[voidptr]

The Falcon Heavy is a 2.5 stage design. The two side boosters will stage before the center core separates; it'll land as three independent cores.

Re:Larger landing area

[towermac]

Yeah.

I gotta think that there is some value in the barge, beyond the showoff value of landing his rocket in spite of the fact he hasn't gotten his permit yet. Obviously, consecutive landings will greatly speed up the application process.

And he wants to build a thing in Texas too. It occurs to me that he might want to launch his bigger rocket from Texas and land the first stage at Kennedy. The barge is good practice for that using the Falcon 9, and he can get his data just as well. See, he knew he was going to have to crash a few to get it right. Where better to crash the first few than out in the Atlantic. Far bigger pain in the ass to clean up the molten and burning rocket, whose smoke could be seen from half the state, at Kennedy.

So perhaps that in itself is enough reason to go with the barge. But Elon is the kind of guy that is 3 moves ahead, not 1 or 2. Consider that he will be able to sell a launch to anyone.

I mean anyone. All the client has to do is to pour a good hard concrete slab, and they can launch the rocket from right there, wherever they are. No need to get your Ministry of Whatever to approve a rocket landing; Elon takes care of all that himself with no muss or fuss. No need to consult the US or NASA or the EU of schedule time somewhere...

The nation of Tonga can now have a dynamic and robust space program. It will be that cheap.

Rent-A-Rocket is going to be fucking huge, and I think the barge will be a big part of it.

Re:Larger landing area

[Anonymous Coward]

"I'd piss on a spark plug if I thought it would do any good!" - General Beringer

Landing vs splashdown

[magarity]

One would think lifting off with all that fuel needed for the landing is inefficient compared to a splashdown parachute recovery like the shuttle's boosters. And the damage caused by landing on water with parachutes has got to be less than the explosions from the landings on the barges.

Re:Landing vs splashdown

[DaHat]

And the damage caused by landing on water with parachutes has got to be less than the explosions from the landings on the barges.

Probably not when they figure out how to land on the barge without exploding... at that point the damage from hitting the water and amount of cleaning & service required to be read for launch will be much more.

Re:Landing vs splashdown

[Anonymous Coward]

The engine might be the difference. A water landing would damage a lot of the components while a successful boat landing would leave them intact. If the cost of rebuilding the engine after a water landing is more than the cost of extra fuel needed to facilitate a barge landing then the barge landing makes a lot of financial sense.

Plus it's just pretty damn cool to be able to land a rocket like that.

Re:Landing vs splashdown

[Bruce Perens]

They have talked about refueling on the barge and flying the booster to land! That's really difficult to do after a salt-water dip :-)

Re:Landing vs splashdown

[Anonymous Coward]

It might be. I guess it just makes more economic sense to reuse the first stage and pay for a bit of extra fuel than the splashdown.

I would assume the first stage probably is less critical in terms of fuel because it doesn't have to go that far so you are only carrying all that extra fuel for a little while on your trip.

Re:Landing vs splashdown

[Anonymous Coward]

One would think that if splashdown parachute recovery was the alternative. It's not.

Re:Landing vs splashdown

[tquasar]

Salt water would likely cause too much damage. Go park your car in the ocean for thirty minutes and let me know how that works out.

Re:Landing vs splashdown

[dpilot]

I've heard this point before, with the obvious comparison of Shuttle wings. The counter is that wings are absolutely dead weight on liftoff, plus you've added an entirely new structural mode to the airframe. It has to have the correct structural strength for both vertical ascent and horizontal landing. Both wings and bimodal structure add weight.

Landing the F9 on it's tail, it's practically empty, a fraction of it's initial weight. I'd be interested in seeing the math between F9 and Shuttle, but I suspect SpaceX has done their homework on this.

Of course the science fiction idea of landing anything that can then take off is just that - science fiction. The LEM did it, but then again, only half of the LEM - the bottom was left behind.

Re:Landing vs splashdown

[Ichijo]

I wonder if they've considered adding airbags to help cushion the fall, protect the rocket from water damage, and provide flotation?

Re:Landing vs splashdown

[Beck_Neard]

Every time there's a thread about this, someone says the same stupid thing you're saying, and it's still wrong.

There's a huge difference in power requirement in getting a fully fuelled and loaded rocket up in the sky, and slowing the descent of a nearly-empty, lightweight fuel tank. You need very little fuel to accomplish the latter. Don't forget that parachutes have mass too and it's very hard to make a controlled descent with them (especially if you need to carry the rocket a significant distance). All in all, the solution that best combines cost-efficiency with the ability to land precisely is the vertical retro-rocket landing that SpaceX adopted.

Re:Landing vs splashdown

[Calinous]

They have fuel to spare on board anyway (so they can launch with eight of nine engines working)

Re:Landing vs splashdown

[Bruce Perens]

Where I used to live they had trucks dump salt on the road!

Re:Landing vs splashdown

[AaronW]

A couple of months ago I was having a discussion with a fellow from Space X who designs the hydraulic systems and we spoke about a number of issues. This was right after the failed landing due to it running out of hydraulic fluid. I asked about how reusable the engines are and he said that they run test burns lasting hours. The launch is only a few minutes. According to what he said, it should just be a simple matter of refueling and adding more hydraulic fluid and probably some other simple things without having to do a major overhaul. The engines are very reliable.

I asked about why they don't reuse the hydraulic fluid and he said that it was cheaper and lighter to not reuse it. He also said that they knew it could run out and that the next version would have more.

Re:Landing vs splashdown

[DrunkenTerror]

Is it a Toyota rocket? Be fine.

Re:Landing vs splashdown

[Anonymous Coward]

Hoy aboyt a net

Re:Landing vs splashdown

[Bruce Perens]

We could start with our already phallic looking rocket and then have it come down into something that looks like the world's largest inflatable sex toy. Elon Musk might have trouble living that one down. :-)

Yes, there have been many proposals to somehow catch the rocket.

Re:Landing vs splashdown

[halltk1983]

Hard to splash down on the moon, Mars, asteroids and just about everywhere else we want to go. We'll have to get it right eventually, might as be now. Bonus benefit: cheaper than overhauling the engines every time. You'd think with them doing this at a third the cost of anyone else, WITH A PROFIT, that people would understand that they know what they're doing. Yes, there will be early failures, but this doesn't add that much cost, especially considering long term payoff.

Re:Landing vs splashdown

[Bruce Perens]

It's lighter to not reuse the hydraulic fluid.

It's an open loop system with pressurized gas pushing the fluid out and then it's dumped in the air. Pumps and whatever powers them have weight.

Remember that big fuel-required multiplier in getting any weight at all to 78 miles height and suborbital speed.

Re:Landing vs splashdown

[kupekhaize]

If one engine fails, the fuel cross feeds from that fuel tank into the other engines. I could be wrong, but I do not believe each fuel tank automatically includes more fuel to burn that particular engine for longer just in a case another engine fails. That would be an awful lot of redundant weight for very little benefit. Being able to cross feed the fuel from a broken engine's tank into the others would make far more sense then constantly launching with a bunch of dead weight that is not going to be used. Yes, the xfeed gear needed to transfer fuel from one tank to the other is also extra weight, but it's not like launching with a bunch of heavy fuel that isn't going to be used.

Re:Landing vs splashdown

[Bruce Perens]

The first stage is most of the rocket by weight. The second stage has one engine, while the first has 9. And as you can see from a photo, the second stage is much smaller.

What makes it recoverable is that it doesn't take much fuel to bring an empty first stage back down. It's really light when empty. They only use one engine out of the 9, for very short burns, to do that.

Re:Landing vs splashdown

[Anonymous Coward]

Salt water would likely cause too much damage. Go park your car in the ocean for thirty minutes and let me know how that works out.

Whoosh :-) Mr Quasar, I think that point was made with "That's really difficult to do after a salt-water dip :-)"

Re:Landing vs splashdown

[Anonymous Coward]

I assume this is a rhetorical question. I'm guessing the folks who were bright enough to build it in the first place were bright enough to include n *all* of the issues into the cost-benefit analysis.

Re:Landing vs splashdown

[Bruce Perens]

The required longer burn to make up for a shut down engine does use extra fuel. It also changes the orbital injection point.

Re:Landing vs splashdown

[Dunbal]

Consider it deducted from the amount of payload/second stage you can carry. But if your first stage is engineered to be more than enough for what you need as payload anyway (ie it's already cost effective) you might as well try to recover the cost of it with a soft powered landing and make MORE profit.

Re:Landing vs splashdown

[Dunbal]

You're adding more mass, which requires even more fuel.

Re:Landing vs splashdown

[Anonymous Coward]

Less thrust -> more gravity losses.

Re:Landing vs splashdown

[Anonymous Coward]

parachutes and airbags get your rocket wet with corrosive seawater and smashed...not much good if you want to relight it after refueling without a piece by piece refit...

Re:Landing vs splashdown

[beelsebob]

Yes, there's fuel cross feeds, but with lower engine power the ship sticks less close to the optimal acceleration curve, and hence needs a longer burn.

Re:Landing vs splashdown

[holmstar]

But you don't fill the engine and passenger compartment with saltwater.

Re:Landing vs splashdown

[Mr D from 63]

Probably not when they figure out how to land on the barge without exploding... at that point the damage from hitting the water and amount of cleaning & service required to be read for launch will be much more.

Maybe landing on a platform is not the best approach. Rather, I seems like some type of gantry that could grab the top of the rocket when it is just above the water would be easier and more reliable than having the rocket land and maintain its own upright position.

Re:Landing vs splashdown

[bill_mcgonigle]

fuel needed for the landing is inefficient compared to a splashdown parachute recovery

The barge/ocean is just a temporary measure. The vision is twenty rockets launching a day and returning to the launch site to prep for the next launch.

There were about 120 rocket launches last year. SpaceX's mission statement is to reduce the cost of launches by 100x, and utilization rates go up as costs fall, so it's not just 100x more launches - twenty a day is probably very conservative if they hit their price targets.

Queue the folks who can't imagine what anybody would do with more than 640 launches a year.

Re:Landing vs splashdown

[AaronW]

Which is basically what I said in my second paragraph. It was a good talk. It's not everyday when you can sit down and talk with a rocket scientist in a casual environment.

Re:Landing vs splashdown

[Applehu Akbar]

Remember: seawater ruins everything.

Re:Landing vs splashdown

[Applehu Akbar]

I remember road salting also. And every car more than about two years old has huge suppurating rust holes in it.

Re:Landing vs splashdown

[Anonymous Coward]

I can imagine a tremendous uptick in global temperature with all that delicious extra heat pouring into our atmosphere.

So long as Elon gets ever richer, I'm all for it.

Re:Landing vs splashdown

[camperdave]

It's not everyday when you can sit down and talk with a rocket scientist in a casual environment.

... or a brain surgeon.

Re: Landing vs splashdown

[Anonymous Coward]

Yeah, repeating the achievements of the 1950s must be impressive to fanboys.

Re: Landing vs splashdown

[Ukab the Great]

Like most pieces of hardware, the phallic rocket will be on top of your busy box.

Re:Landing vs splashdown

[Bruce Perens]

Sorry.

I guess then you were not so lucky as to have rocket scientists in the family. I guess I'm not unlike many techies my age, whose dads worked in aerospace. My dad worked on the lunar module at Grumman. My father in law worked in the blue cube for Lockheed.

People think of me an the Open Source guy. But I have been getting space spoon-fed to me since before first grade.

Re:Landing vs splashdown

[Anonymous Coward]

If they were gonna fly the booster to land why not just land on land in the first place.

Re:Landing vs splashdown

[kabir_nk]

Not more intrest on it

Re:Landing vs splashdown

[Anonymous Coward]

I mean look how trashed this thing is.

Re:Landing vs splashdown

[AJWM]

One would think that if they didn't know that the shuttle's boosters are made of inch-or-more-thick steel, while the Falcon's tanks are millimeter thick aluminum-lithium. And that the booster splashdown still tended to leave the boosters slightly out of round (which contributed to the problem Challenger had).

The extra fuel almost certainly weighs less than the necessary parachutes would.

Re: Landing vs splashdown

[jordanjay29]

You think landing on a body 1/6 of Earth's gravity, without an atmosphere or weather, and under the control of a human is really comparable to landing on Earth, with full gravity, atmospheric weather systems, and all controlled by a computer?

Re:Landing vs splashdown

[AJWM]

DC-X also did it, several times -- but then DC-X wasn't trying to make even a fraction of orbit, it was proving the vertical takeoff and landing principle. Its engines (modified Pratt & Whitney RL-10s) could be more deeply throttled than the Falcon's Merlin, and it (the DC-X) was built fairly heavy to start with, since was designed as a test vehicle rather than a launcher (fully-fueled the legs couldn't hold its weight, it needed a support structure for takeoff -- and in an abort (happened once) it had to hover until it had burned off enough fuel to land).

Since then a number of small-company-built test vehicles have done the same, although not (afaik) to the tens of thousands of feet altitudes that the latter DC-X flights made.

Re:Landing vs splashdown

[AJWM]

All the engines on the Falcon 9 (and just about every other multiengine* rocket stage) are fed from the same propellant and oxidizer tanks. Giving them separate tankage just adds weight and plumbing complexity.

In the Falcon Heavy, there is a cross-feed mechanism from the outrigger 9s to the core so that the core can keep burning when the outriggers jettison (saving weight).

*(except multiengine solids, where the engine is the fuel tank.)

Re:Landing vs splashdown

[Anonymous Coward]

Saltwater and rocket parts DONT mix - the SRB's from the shuttle had to have 5000 parts refurbished after every launch. SpaceX's design is so they shouldnt have to refurb anything between flights (or at least it will be minimal in comparison)

Re:Landing vs splashdown

[SuricouRaven]

It's worse than a multiplier: It's an exponent.

Re:Landing vs splashdown

[Anonymous Coward]

This. They are learning while doing something cheaper than has been done before. Slashdot used to be smarter than this.

Re: Landing vs splashdown

[kav2k]

Undoing fat-fingered moderation, should be +1 Insightful

Re:Landing vs splashdown

[AaronW]

My father was is engineer who worked on some aerospace and defense stuff while I was growing up. He worked on a data recorder that went up on Skylab. He offered to fly up and fix it if anything went wrong. My mother has always been big on astronomy. I've been Elon Musk's various ventures for quite a while (and ended up buying a Tesla). I've always been impressed with his forward thinking and the smart people he surrounds himself with. My day job is working on the U-Boot bootloader and a little Linux kernel work. Maybe someday it'll even be pushed upstream, but right now it's always not enough time and it's a LOT of code, significantly more than all of the ARM SOCs combined.

Re:Landing vs splashdown

[Simon Brooke]

Remember: seawater ruins everything.

One of those occasions where I wish I had mod points but don't. Mod the parent post up!

Seawater is extremely corrosive. Engineering the rocket engine to survive sudden immersion in seawater when very hot would add a great deal to the complexity and cost (and probably weight). And that's before you add the cost of engineering the rest of the vehicle to resist corrosion.

Re:Landing vs splashdown

[Simon Brooke]

Close to where I live are large intertidal mudflats. Every other summer some tourist drives a brand new four by four out there and gets stuck. And then, of course, the tide comes in. When the vehicles are recovered two or three tides later, they are insurance write-offs - the electrics, interior, and engine are all beyond repair.

You do not want to immerse something complex and expensive in salt water unless you really, really have to.

Re:Landing vs splashdown

[SkyratesPlayer]

Grabbing it would almost certainly crumple the tank. It is not designed to take much in the way of sideways loads.You would have to add a flange or some hooks to the top of the first stage, and thicken the skin, which would add significant weight (although getting rid of the legs would help). The elegant thing about a pad landing is that it is just like a launch in reverse - all the loads are in the same direction. It's just that the mass is way less and the CG is much farther down.

Re:Landing vs splashdown

[Anonymous Coward]

You're welcome to stay here while the rest of us jaunt off to ALPHA CENTAURI for the PARTY OF THE MILLENNIUM!

Re:Landing vs splashdown

[Anonymous Coward]

Huh? They salt the roads every year here. And we don't have a huge problem with rusting cars.

Re:Landing vs splashdown

[Mr D from 63]

Good points. thanks

Re:Landing vs splashdown

[Anonymous Coward]

Not fish.

Re:Landing vs splashdown

[Andy Dodd]

I live in one of those areas, which is why older vehicles that are still on the road in respectable quantities in Southern states simply don't exist here... Even with decades of improvement in corrosion control/resistance, vehicles in "rust belt" states like New York don't last nearly as long as the same vehicle would in a place like Georgia or the Carolinas.

Re:Landing vs splashdown

[Andy Dodd]

"He also said that they knew it could run out and that the next version would have more." - Hence Elon's comment back then that the next one should go boom for a different reason. He was right. :)

Re:Landing vs splashdown

[edremy]

You mean how trashed the space is where the old Constitution is? It's basically George Washington's axe at this point- there's very little left of the original ship, just all the stuff that they patched her with. She was overhauled and rebuilt multiple times while still on duty, but the article you sent notes that by 1925 she was in such bad shape the stern was ready to fall off and had to be fully repaired. She was repaired again in 1973, then in 1995 they rebuilt her again to return her to her original design.

Re: Landing vs splashdown

[Anonymous Coward]

I'd say that the biggest problem here is that the F9 is not optimized at all for landing - if the engines could throttle down to close to 0%, and there were enough small rocket engines spread around the top for attitude control, landing would be much easier. Since the former would probably make the engines more complex or less efficient, and the latter would just add significant dead weight on the way up, SpaceX has apparently chosen to try without.

Re:Landing vs splashdown

[Bruce Perens]

Remember the reporters asking what was holding DC-X up? They couldn't see the rocket exhaust.

I sneaked inside of the Rotary Rocket the last time I was in Mojave. Someone left the bottom hatch ajar. But there wasn't a way to climb up to the cockpit from in there. Lots of pigeon droppings and it's used to hold the equipment for the multimedia kiosk nearby. Sad to see.

Re:Landing vs splashdown

[rpstrong]

I had a '69 Merc "woody wagon", where the wood sides were basically wood grained Contact paper. By the time I owned the car (in '76'), that plastic had become structural . . .

Re:Landing vs splashdown

[tquasar]

I camped near San Felipe in Baja California and a dummy was driving his car along the beach and got stuck in the sand at low tide. The tides are from eight to twelve feet in the Sea of Cortez and his vehicle was covered by water twice until a local arrived with a mini Caterpillar to pull the vehicle from the sea.

Re: Landing vs splashdown

[tibit]

The moon landing was with human-in-the-loop, but it was automatically stabilized and would have been impossible otherwise IIRC. Had the automation failed, they'd have to abort back to orbit. An automatic controller kept the thrust vector going through the vehicle's center of mass. The humans controlled the orientation and thrust level.

Re: Landing vs splashdown

[tibit]

Sorry, that should have read: humans controlled translation, rotation and thrust level. The automation also decayed the roll rates to zero, so that if you'd reorient, it'd stay that way by itself.

Re:Landing vs splashdown

[toddestan]

The quality of the paints and rust-proofing has improved immensely in the past 20 years or so. Back in the 70's in salt country it was common for 2-3 year old cars to have rust, and at 7-10 years cars were commonly scrapped due to major structural issues. Nowadays, it's rare to see any rust on a car less than 10 years old (unless it's a Mazda) and even for 15-20 year old cars the rust is still mostly cosmetic.

Success, as proof of concept

[Anonymous Coward]

Or are the rules inflexible in this matter

No I don't agree

[MichaelSmith]

The falcon can't throttle down enough to hover before landing so it has to approach the pad at high speed, and high acceleration. While doing this it has to rotate the entire vehicle to control lateral movement. It has to coordinate lateral and vertical acceleration to achieve near zero in all three axes at touchdown, with only one chance to get it right.

I doubt this can be done without extra thrusters for fine control over velocity and position.

Re:No I don't agree

[Bruce Perens]

But we've seen Grasshopper and Falcon R9 position properly on land. Nobody's told us what the maximum wind was in those tests.

Musk alluded to a process control issue this morning and then deleted the tweet. It will take time to find out what the deal is.

Merlin 1D can throttle to 70% and the old 1C could go to 60%. Perhaps there's room for deeper throttling. I would expect that they'd try that before adding a new system and its weight.

Re:No I don't agree

[Bruce Perens]

In those tests, both Grasshopper and R9 were coming down much more slowly. But it appeared they could come down slowly. Pretty close to hover.

Re:No I don't agree

[stox]

I suspect they just need to tweak some PID loops at this point. I don't think throttling is an issue. Unlike the Lunar Lander, the target of the landing is a concrete point. Hovering should not be needed. Drop it on the spot, done.

Re:No I don't agree

[Immerman]

>Perhaps there's room for deeper throttling.

Indeed. It might even make sense to use a different design for the central engine if deeper throttling interferes with efficiency or maximum thrust. Of course if they can get the control systems working properly with the standard engines that would be the optimal solution.

Re:No I don't agree

[Pope Hagbard]

Came here looking for the armchair rocket scientists, left unsurprised but disappointed. Dunning-Kruger lives.

Re:No I don't agree

[bill_mcgonigle]

But it appeared they could come down slowly. Pretty close to hover.

Yeah, I think that's the inevitable end-game - there is plenty of time to make small adjustments right up until the point of contact with the solid parts of the planet. AIUI, they're so close to empty on the fuel tank after the burnback that they're trying to get it down on the pad ASAP. They can only attempt these landings for now on launches that don't require as much fuel as others - supposedly the next iteration can hold more fuel.

Re:No I don't agree

[Anonymous Coward]

PID: explain.

Yeah, I didn't think so. But you sound so reasonable mouthing words you have no understanding of.

Re:No I don't agree

[Anonymous Coward]

Bro, they just need to PID the sticktion control algorithms to wobble the hydraulic line friction out of phase with the control surfaces of the launch vehicle.

I parroted a bunch of buzzwords I saw in a SpaceX press release - I think that demonstrates that I know exactly what I'm talking about.

Re:No I don't agree

[Anonymous Coward]

I suspect they're using full state optimal adaptive control and not just a PID loop.

Re: No I don't agree

[djdarko]

An upright rocket being controlled by a mixture of aeronautical control surfaces, a gimbaling, throttleable rocket engine, and thrusters is majorly non-linear and also time-varying, especially when you factor-in the effects of fuel burn, vehicle attitude, and huge changes in velocity on the system's dynamics. There is no way that they are using a simplistic linear controller (like PID) to pull this off. I suspect that they are doing some of the most cutting-edge adaptive / non-linear control design in the world. Hats off.

Re:No I don't agree

[SuricouRaven]

We have Kerbal players now.

Re:No I don't agree

[gweihir]

Well, basically they are aiming to get some degree of re-use. If, say, 4 in 5 can eventually be recovered, that would be significant cost savings. Seems they still have quite a while to go, but that is actually normal at this stage.

Re:No I don't agree

[cjameshuff]

Those test vehicles weren't coming back after launching a second stage on its way to orbit, and could carry a bunch of extra ballast. For the later hoverslam tests, they still lifted off with a T/W ratio of 1, hovered for a bit to burn off fuel, then started their descent under power before their T/W went above 1. The returning first stage starts its landing burn with a T/W ratio of more like 2-2.5.

Re:No I don't agree

[slew]

Proportional Integral Derivative (a very primitive type of feedback loop controller that modifies the control input a fraction proportional to the error signal, another fraction relative to the integrated/smoothed error signal, and yet another fraction relative to the first derivative of the error signal). In the old days people spent lots of time "tuning" these fractions (e.g., like ziegler nichols) to get acceptable performance.

I suspect they don't use anything this primitive in their control loop (at least I hope they don't). That's control-systems 101 in college before you get to smith controllers and state based systems feedback (which is also quite primitive)...

Re:No I don't agree

[tibit]

The "rotation of the entire vehicle" part was because the scenario, currently, is:

1. Use aerodynamic lift and grid fins to aim to the side of the barge.

2. Start the engine.

3. Re-aim to the middle of the barge.

They do this since they still fear that a relight might fail and the heavy octaweb structure will punch a hole through the barge. It may well be that the next landing will be aimed at the bulls eye already during the final, unpowered part of the descent.

I doubt this can be done without extra thrusters for fine control over velocity and position.

It can be. We've already seen it done. The difference between success and failure in this most recent case has been very slight. It doesn't take any major redesign to fix that sort of a thing. Get real.

Re: No I don't agree

[tibit]

An aside: If you really, really wished to, you probably could use something as simplistic as a PID, except that there'd be a nonlinear, time varying transform applied to its inputs, and its outputs. At that point you can pretty much almost implement any controller you wish while still claiming that there's a PID in there somewhere :) Think of the Park and Clarke transforms used in motor controllers. The overall controller still uses a PID core, but behaves nothing like a raw PID would. The transforms could be much more complex, effectively implementing a whole different control scheme, and letting PID not do much of substance :)

Re: No I don't agree

[djdarko]

The Park and Clarke transforms are very useful as a means to render a regularly time-varying system stationary by effectively putting the controller in a rotating frame of reference. This does not work in the case of a rocket where the time-varying nature of the system is not predictable or regular like it is with AC electrical systems. They could use a technique such as compensator scheduling that is effectively a set of a linear controllers that are smoothly switched between as the system's operating point changes. This assumes, however, that the operating point variation is slow compared to the system dynamics that the controller is attempting to regulate, which is highly unlikely in this case. And a controller like this is only locally linear and is actually fundamentally non-linear.

Try HD mode

[Bruce Perens]

You can see a lot more if you go to 1080 HD and full screen. There's some large piece of equipment, perhaps the motor head for one of the barge's corner thrusters, being thrust off of the barge in flames.

It looks like they'll need to do a lot of work on the barge. The support ship Go Quest and the tug Elsbeth III seem to be back in Jacksonville according to vessel tracking sites. There is a Carnival cruise ship that parks next to the barge's dock every 4 days, so we will probably see photos from its bow netcam if we don't see them otherwise.

Oh, check out this newscast. At 2:43, CBS News uses a sequence a SpaceX fan produced with Kerbal Space Program to illustrate how the landing is supposed to work.

Re:Try HD mode

[Anonymous Coward]

At 2:43, CBS News uses a sequence a SpaceX fan produced with Kerbal Space Program to illustrate how the landing is supposed to work.

You mean CBC?

Re:Try HD mode

[Bruce Perens]

Oh, they're Canadian. That explains everything!

(Ducks and runs for cover)

:-)

Re:Try HD mode

[Vrallis]

I don't think that was Scott Manley's video, but he's hardly the only well-known person making KSP videos. Given the effort it takes to recreate it that well I doubt someone on their staff did it. I wonder who they ripped off without credit if that's the case.

Re:Try HD mode

[dbIII]

More posts like that and we'll mistake you for ESR again.

Re:Try HD mode

[Samizdata]

I concur, especially as the Bandicam watermark shows it is an unregistered copy of the recorder software, and I suspect using such would mean bad actionable things for CBC News.

Close? Really?

[Anonymous Coward]

After watching Armadillo go from better-luck-next time, to close, to closer, ...., ...., ...., oh so close, to wow that was so close; this doesn't seem all that close... Not at all...

The Hard Way

[rtb61]

Why do it the hardest and most fuel inefficient way imaginable. Split seam the fuel tank, then swing out and rotate the elements and create a massive autogyro. Sure you have to be careful with the seal of the seams in the tank and wind could be a problem requiring on the fly change of landing zone but overall a whole lot less additional fuel required and even a bad landing will still be soft by comparison.

Re:The Hard Way

[Bruce Perens]

Split seam the fuel tank, then swing out and rotate the elements and create a massive autogyro.

They have a job for you in the ULA marketing department.

Re:The Hard Way

[gstoddart]

Split seam the fuel tank, then swing out and rotate the elements and create a massive autogyro

Why am I not convinced your way sounds like the "easy way"?

I can't event think of the mechanical stresses involved in opening this thing up to spin it around.

In fact, it sounds outright crazy.

And that's before we start considering a fuel tank designed to open up. Because, what could possibly go wrong there?

Re:The Hard Way

[viperidaenz]

What's the worst thing that could happen? The fuel tank leaks and explodes shortly after takeoff, showering the ground nearby with burning wreckage, or leaking rocket fuel higher up in the atmosphere, without igniting it, coating everything around with toxic, un-burnt fuel, ready to set half a state on fire?

Re:The Hard Way

[Immerman]

Who cares about the fuel efficiency of landing? A Falcon 9 launch costs about 50-56 million dollars - the fuel itself only costs about 200,000, or 0.4% of the launch costs. The cost of increasing the fuel load by a few percent to allow for landing barely even qualifies as a rounding error.

Meanwhile, all the extra mechanisms needed to turn a simple high-strength fuel tank into a transformer is almost certainly going to increase the mass so much that you'd need far more fuel to get the thing to the second-stage separation point in the first place - so net loss on fuel anyway.

Re:The Hard Way

[Anonymous Coward]

Relax; it's only kerosene.

Re:The Hard Way

[itzly]

Who cares about the fuel efficiency of landing?

Don't forget that you need to launch all that fuel to a suborbital trajectory first, this requires a bunch more fuel. The mass of all that extra fuel comes from your payload budget.

Re:The Hard Way

[khallow]

Why do it the hardest and most fuel inefficient way imaginable.

There are two things to note here. First, it's not the most "fuel inefficient" way as your alternative demonstrates (all that crap has to go up as well as come down, that uses far more fuel than the current approach does). Second, fuel efficiency is not that important. The rocket engines that SpaceX is trying to recover are far more costly than the additional propellent required for the current scheme to recover them.

Re:The Hard Way

[itzly]

The cost of the fuel is not a factor, but the mass is.

Re:The Hard Way

[khallow]

But that mass is not as big a factor as recovering those engines. It's not just not having to build a new set of engines every launch, but also reliability. You have both tested engines in actual flight and some increased ability to recover failed rocket engines.

Re:The Hard Way

[itzly]

But that mass is not as big a factor as recovering those engines.

But maximizing payload mass is very important, since it's already a small fraction of total rocket mass, and it's paying for everything. Every pound of extra fuel you want to keep for landing is coming directly out of your payload budget.

So, yes, recovering the first stage is very important, but doing it in a mass optimized way is just as important.

Re:The Hard Way

[cjameshuff]

But maximizing payload mass is very important, since it's already a small fraction of total rocket mass, and it's paying for everything. Every pound of extra fuel you want to keep for landing is coming directly out of your payload budget.

No, it comes out of the entire second stage mass, only about 20% of which is payload/structure. With that mass fraction, each kilogram of payload you sacrifice means there's 4 kg of second stage propellant you no longer need, so you can load an extra 5 kilograms of propellant onto the first stage. They anticipate around a 30% payload reduction due to reuse, eliminating most of the cost of a new launch vehicle in return.

Re:The Hard Way

[Immerman]

Sure, that's why they don't attempt re-use on launches near the limits of the Falcon's capabilities. But it's not like they can just build a smaller rocket for normal launches, so as long as they have the unused lift capacity carrying a bit of extra fuel comes almost for free. And it's still nothing compared to substantially increasing the mass of the Falcon itself - the extra fuel needed in that scenario *also* requires a bunch more fuel to lift - face it, it's a losing proposition. Even a simple parachute that could support that kind of weight is probably going to mass more than the extra fuel needed to do the same job.

Re:The Hard Way

[rtb61]

Yet, you managed quite foolishly ignore risk, right in the face of the self evident consequence of it. It goes wrong, it goes boom and you have nothing. Besides staged rockets logically become a thing of the past, when you want to establish a permanent moon base (they become parts for the base and for vehicles that will go further out into space).

Re:The Hard Way

[rtb61]

All of it, or part of it, what designed rate of rotation and before or after a parachute is deployed to get it to design deployment speed, of course the tank is now a empty vessel by far the majority of mass is gone and how well the balance between centrifugal action and air pressure is handled in deployment. Banged up, is always better than blown up.

Re:The Hard Way

[cjameshuff]

Yet, you managed quite foolishly ignore risk, right in the face of the self evident consequence of it. It goes wrong, it goes boom and you have nothing.

If you throw it away, you are guaranteed to have nothing. If you bring it back for reuse, you have a great deal of very expensive hardware that you no longer need to build for the next launch. Just reusing it once allows that hardware to launch 50% more mass over its lifetime than it would have if operated as an expendable vehicle. They anticipate much longer lifetimes.

Besides staged rockets logically become a thing of the past, when you want to establish a permanent moon base (they become parts for the base and for vehicles that will go further out into space).

First...that doesn't in any way make staged rockets "a thing of the past". Second, the first stage doesn't go anywhere near low orbit, let alone the moon.

Re:The Hard Way

[rtb61]

Of course there is always doing away with the first stage or more accurately leaving it on the ground, the BB gun approach. Take a long tube, under vacuum, insert launch vessel under magnetic support and release compressed air behind the launch vessel, to accelerate it up the tube to achieve desired speed and the highest reasonable launch altitude. Now to improve the qualities of the compressed air, use hydrogen and oxygen and ignite it, within the tube behind the launch vessel. A really long tube with multiple ignition points, which kick off once the launch vessel has passed, to ensure desired pressure is maintained. Basically an extended ignition gun with much reduced shock and acceleration distributed across the length of the barrel. Small, low altitude satellites a speciality.

Moving platform?

[cdxta]

Came in to hot and didn't account for thrust pushing barge down?

Errr

[sunking2]

Close as in a clock is right two times a day. I applaud the attempts, it makes great video. But am not convinced this is the right course of action.

Re:Errr

[Anonymous Coward]

But am not convinced this is the right course of action.

What does that mean? Would you like to explain to the class why the rocket scientists are following the "wrong course"?

I Disagree with the Summary

[cruff]

Looking at the video, it appears the booster does not come close to ever having anywhere near a true vertical orientation, and this attempt was not, in fact, "incredibly close to success". Granted, it came closer than ever in history to achieving the goal, but the thruster appeared to not have enough thrust to push the rocket to a vertical position once the booster touched down on the barge. I hope Space-X has a successful next test! The world needs a dose of rockets landing on large flames in the style of those old campy movies.

Re:I Disagree with the Summary

[Michael Woodhams]

I was shocked at how abrupt and extreme the pitch changes were. I think so long as it needs such gross adjustments so close to landing, landings will be unreliable with a significant chance of failure. It is not at all like the tidy landings made by the Grasshopper test vehicle.

Two engineering changes which could make a big difference are lower minimum thrust (so it can approach the landing with lower acceleration) or lateral control rockets (RCS) at the top of the stage.

Re:I Disagree with the Summary

[Anonymous Coward]

They hit a barge in the middle of the ocean with a gigantic rocket that was nearing orbital velocity. I think we need to cut them some slack :)

Re:I Disagree with the Summary

[adolf]

It seemed to have plenty of thrust, to me: The amount of deceleration was remarkable.

Meanwhile, it's not a matter of getting the rocket vertical after touchdown, but before: Once it "touches down" with one or more of its feet, the dynamics instantly change (surface friction becomes a thing, and a tall free-floating object grows a fulcrum) and all bets are off.

Go play some Lunar Lander and try again. :)

Re:I Disagree with the Summary

[beelsebob]

If you watch the video, you'll notice that it does in fact have RCS thrusters at the top of the stage.

Re:I Disagree with the Summary

[Michael Woodhams]

Someone in the Youtube comments says "The flight profile veers the booster off to the side on purpose so the exhaust from the final burn isn't directed at the barge where it could do damage"

If this was a planned manoeuvre, I'm much happier. Can anyone confirm this statement?

Re:I Disagree with the Summary

[Zaiff Urgulbunger]

^ this

Re:I Disagree with the Summary

[cbhacking]

It *has* RCS at the top of the stage. One of them fires for a good 4-5 seconds trying to hold the stage upright after the touchdown (it failed, obviously). Were you watching in really low quality or something?

Re:I Disagree with the Summary

[Anonymous Coward]

This looks a bit like See Harrier landing, where it first hovers over the ocean, and then hops to the carrier. From my very limited experience with aircrafts (quadrotor owner) , hovering close to the surface is very problematic because air you push down reflects back and affects the aircraft, preventing you from doing any meaningful steering. Plus if you are flying a jet/rocket engine, you may burn the platform and aircraft.

Re:I Disagree with the Summary

[Henriok]

If this is true, I haven't seen any of their test vehicles doing this in try outs. They all hovered very controlled, and descends slowly and steadily. This landing was from what I could see the definition of "out of control", trying to regain control and desperately trying to complete the mission.

Re:I Disagree with the Summary

[Anonymous Coward]

It come in leaning right, it falls over to the left, it must have hit vertical some where in between those two states...

In high def you can see the Draco thrusters on top of the first stage trying to push it back up right as it lands... That thruster is only capable of 90 pound of force... and it holds that rocket up for several seconds.

Swap out the Draco's for Super Draco's and those thrusters could have held it together easily... A mid tier Draco is probably in order... don't need anything as big as the super Draco's (which were designed for a retro burn to place a dragon capsule on mars) so a enhanced Draco or a Draco V2 should be able to pull it off nicely.

Re:I Disagree with the Summary

[Anonymous Coward]

Of course https://xkcd.com/695/

Re:I Disagree with the Summary

[Anonymous Coward]

Closer than ever in history? I'm pretty sure NASA demonstrated it in the 80's on earth, and in the 60's on the moon.

Re:I Disagree with the Summary

[Anonymous Coward]

But I've seen rockets do that for decades already. Only they had their tip pointed the other way.

Re:I Disagree with the Summary

[tibit]

I was shocked at how abrupt and extreme the pitch changes were.

They were a part of a pre-planned hole-punch-avoidance maneuver. It was a part of the plant. Imagine what would happen if the engine didn't restart, with the rocket aimed at the middle of the barge. Hence you aim outside the barge, start the engine, then re-aim to the X. I bet that they trust the engine enough by now that they'll get rid of that maneuver in the next landing or two.

"Close" Only Counts

[l0ungeb0y]

With Horseshoes and Hand Grenades. We've seen what "close" gets us with rocketry, and it's not pretty.

Re:"Close" Only Counts

[Anonymous Coward]

With Horseshoes and Hand Grenades.

And atom bombs...

Re:"Close" Only Counts

[Immerman]

In combat it's called "walking your fire". Sure, the first 30 rounds aren't close enough to matter in themselves, but they give you enough information so that the 31st round hits home.

Re:"Close" Only Counts

[Bruce Perens]

I must confess that most of my programs have bugs the first time I write them. I don't start over from zero when that happens.

The Wright Flier didn't get to San Francisco, but it started the path that led there. Actually touching down on the planned point, at the planned vertical velocity, is pretty good. They'll fix the rest.

Re:"Close" Only Counts

[Pope Hagbard]

Right? If we listened to some of these posters we'd never advance from whenever a given poster was a small boy.

Re:"Close" Only Counts

[Anonymous Coward]

"Close" was good enough for Apollo 11

https://en.wikipedia.org/wiki/Apollo_11#Lunar_descent

Re:"Close" Only Counts

[Bruce Perens]

If you think that's bad, read some of the comments to nontechnical news site articles on the recovery failure. Ignoramuses whining "how much of my taxes did this failure use". They aren't even smart enough to realize that it launched the Dragon to ISS successfully, and that NASA isn't footing the bill for recovery attempts. It's really enough to kill one's sympathy for the common man.

Re:"Close" Only Counts

[Fwipp]

They certainly haven't.

Re:"Close" Only Counts

[l0ungeb0y]

I was not implying that they should start from scratch. Just that "close" here is not "close" in terms of a workable solution as the summary had indicated they were close to having the problem solved. And in respect to the summary, I meant only that they are still a long ways off from having a reliable vehicle. While I get your point -- I find it disingenuous in this case as it takes a very simple statement to a very radical conclusion and out of context. And because of that, I have had to waste more time explaining something I should not have had to explain in the first place.

Re:"Close" Only Counts

[ArsonSmith]

Not that great with a 30 round magazine.

Re:"Close" Only Counts

[Bruce Perens]

Well, it did what SpaceX was paid for reliably, which was to send the Cargo Dragon up to ISS in an expendable rocket. All of the NASA demo and supply flights they have done have been successful.

Recovery is so far a secondary and private mission of SpaceX, and Musk did say it had less than a 50% probability of success for this attempt (but a 75% to 80% probability of success for the year).

Me, I'm damned impressed that they can bring that thing from 78 miles high and suborbital speed, and touch the landing gear down on the barge at an acceptable descent rate. I think this is pretty good for the second try and they'll nail it soon enough.

Re:"Close" Only Counts

[Bruce Perens]

Armstrong did punch out of an LM trainer before that mission. It crashed and burned. They say that afterward, he went back to his office and quietly did some paperwork.

Re:"Close" Only Counts

[garyisabusyguy]

Armstrong was an exceptional pilot, I read on old NASA report about him regaining control of a space capsule that started spinning before it could kill them. Something to do with a malfunctioning thruster rocket.

Here it is:
"And, make a decision he did. In a rule-breaking move, Armstrong manually disabled the OAMS thrusters and activated the re-entry control system (RCS) thrusters to stabilize the spacecraft. With hand controllers aboard the spacecraft now functioning properly, correct motion of the capsule was restored."
http://www.spaceline.org/fligh...

Yeah, they were 'hands on'

Re:"Close" Only Counts

[khallow]

Just that "close" here is not "close" in terms of a workable solution as the summary had indicated they were close to having the problem solved. And in respect to the summary, I meant only that they are still a long ways off from having a reliable vehicle.

Actually, it is pretty close to solved. They probably just need to improve the control system a little bit.

Second, they have a reliable vehicle by the standards of current rocketry. They aren't competing head to head with a Boeing 747, but vehicles like the Atlas 5 or Soyuz. And a more reliable vehicle is a matter of using the current reliable vehicle a lot to develop the knowledge to build that vehicle more reliably.

Re:"Close" Only Counts

[Anonymous Coward]

It's really enough to kill one's sympathy for the common man.

Sad to say, such sympathy is misguided, unneeded and ultimately unwanted by your target audience. They're too busy stuffing their faces with Cheetos and pizza to give a shit about that there logic thang.

Re:"Close" Only Counts

[TubeSteak]

Yeah, they were 'hands on'

That was a philosophical choice made by NASA: pilots should have as much control as possible over the systems.
Sort of like the design choice made by Boeing to let the pilot override the automated systems and break the airplane if he wants to vs Airbus limiting the max g-forces a pilot can generate.

It didn't hurt that Eisenhower told NASA to only recruit military test pilots for the Mercury program.
While NASA no longer exclusively recruits test pilots, they still make up a large portion of recruits.

Re:"Close" Only Counts

[Immerman]

And if SpaceX weren't already the cheapest way to orbit without reusing the first stage, that might be relevant to the metaphor.

Re:"Close" Only Counts

[SuricouRaven]

The wright flyer flew five times. It crashed five times. It was just simple enough that the brothers could repair it after each 'landing.'

Re:"Close" Only Counts

[Anonymous Coward]

They are not cheapest. They are charging less then it really costs. But Space X is losing money fast and will be out of business soon.

Re:"Close" Only Counts

[Anonymous Coward]

If you think that's bad, read some of the comments to nontechnical news site articles on the recovery failure. Ignoramuses whining "how much of my taxes did this failure use". They aren't even smart enough to realize that it launched the Dragon to ISS successfully, and that NASA isn't footing the bill for recovery attempts. It's really enough to kill one's sympathy for the common man.

The tards? You're the first I've heard to admit sympathy.

Re:"Close" Only Counts

[ArsonSmith]

I'm sure they'll make up for it in volume.

Re:"Close" Only Counts

[Areyoukiddingme]

I'm sure they'll make up for it in volume.

The Coward is a ULA partisan who comes in here to spout lies pretty regularly. They're not losing money. They're profitable on every launch, by a healthy margin. The Coward can't understand how it happens, since the ULA can barely turn a profit when they charge $480 million for the same thing, so he assumes they're taking a loss.

Elon Musk hasn't been so rich that he can run a rocket company for 13 years out of his own pocket. (Probably he is now, because of Tesla stock, but he wasn't for most of that period.)

Alternative Idea for Landing

[PeteJanda]

Kindergarten Question for SpaceX: why not simply put the equivalent of a safety net on the barge, cut the rocket's engines at an altitude of ~10m and let the rocket fall safely into the net? Less fuel, less complexity and less cost.

Re:Alternative Idea for Landing

[Overzeetop]

The forces required are enormous, and even 10m away the rocket thrust would toast most materials. It still has to be caught in a specific orientation to minimize stresses, which means stabilization. As for stopping further, a 10m fall would probably far outstrip the capacity of the structure. (For comparison, more heavily built high power / amateur rockets are designed for touch down forces equivalent to a drop of about 2 meters). The fuel difference is near zero since the full motion of the rocket must be arrested prior to that final "fall".

It also means that the rockets could never land on an arbitrary location, which would be a future goal. Solving it now is a Good Thing (TM).

Re:Alternative Idea for Landing

[onepoint]

While I think your idea would still leave some damage ... I'm thinking similar ... something like a safety loops that snare it at multiple heights once it touches the barge.

Re:Alternative Idea for Landing

[ColdWetDog]

A booster is really a pretty fragile thing. It's designed to be really strong in one axis only. It also has to be lightweight. Grabassing the thing from the sides is going to make for a bunch of expensive scrap metal.

Re:Alternative Idea for Landing

[Anonymous Coward]

They're not trying to land *this rocket*, they're trying to learn how to land rockets on land. Making it easy to go halfway only goes halfway.

This rocket is an expendable test item. And we get a control lag causing penduluming, so that's a fix to apply for the next one.

Re:Alternative Idea for Landing

[Anonymous Coward]

"The forces required are enormous"

Please show your calculations

" and even 10m away the rocket thrust would toast most materials"

Again, do you even know the temperature of the expellant 10m from the apterture of that particular booster, and have you compared it to the melting point of various alloys that could be used?

"It still has to be caught in a specific orientation to minimize stresses"

Are you so intimately aware of the design parameters for the rocket that you know what the stress limits are for various torques and tensions on the vehicle?

"which means stabilization"

Really?

I'll just stop here... please get back to us with your analysis so we can understand why we are all so obviously stupid.

Re:Alternative Idea for Landing

[Anonymous Coward]

Perhaps if you can grab the hard points used at the launch pad.
The goal would be to pick up some more control ability once you get close.
Then transition to barge is fully responsible. (IE, landed)
Sticking the landing at an instant seems a scary sort of plan.

One thing I don't understand is their plan for when the engine exhaust goes close to the deck.
Seems that they could use a flat deck with no protrusions and some holes/ducts to direct the blast.
Aside for helping with blowing things,
with the right shape, they might provide some self centering effect?

It is impressive that they have gotten this far with the leftovers from the primary mission.
Looks like the servo was not ready for wind shifts as it got lower.
But what actually happened could be something different.
In 60's, the fuel slosh in the tanks was a major unexpected thing to add to the servo.
Sure would like to see what's going on inside the control system.

Hopefully, they are gathering enough data to calibrate a sim where they can fully understand the problem and make a robust system.

Re:Alternative Idea for Landing

[onepoint]

there goes that idea.... makes sense.
So no net from below
no snares or loops

How about some sort of air bags that deploy from the side of the deck, just to stable it, don't have to last long, just long enough to keep it centered for a few seconds.

Re:Alternative Idea for Landing

[Overzeetop]

These are gut reactions, based on a career in engineering structures spanning 25 years including 8 of those with NASA directly, 2 with Orbital Sciences Corporation, and 15 in private practice as a licensed professional engineer. I also happen to build and fly amateur (well, high-power, technically and I don't formulate my own propellant) rockets as a hobby.

Re:Alternative Idea for Landing

[Anonymous Coward]

It's so fraglie, that before launch it's held in place by the "strong back" attached to it, which is only removed after the fuel tanks are pressurized, which gives the whole rocket some required extra strength.

Re:Alternative Idea for Landing

[SuricouRaven]

Maybe we could replace the barge with the world's largest box of packing foam peanuts? Non-flamable foam, of course.

Re:Alternative Idea for Landing

[tibit]

Imagine that the net has to catch a bulldozer with a long, empty soda can rigidly attached to it. The weights and relative strengths involved are similar. Your objective is to not crush the can. Now you know why they didn't do it.

Re:Alternative Idea for Landing

[PeteJanda]

Point taken, but a seemingly straightforward answer to the issue you raise is popping the empty soda can off of the bulldozer before both bodies fall into a net. I'm not close to the design and engineering considerations of this particular project, but the current solution strikes me as overly complex and therefore susceptible to a better, simpler mousetrap somewhere down the road.

Any ideas for improvements?

[blackiner]

Watching this got me curious if there were any radical ideas that might help something like this work better. What if you fired a bunch of harpoons that latched on to the rocket as it approached to try to help guide it down? Although they would need to be on some kind of spinning track since the rocket could spin a lot, and any latches added to the rocket would almost certainly mess with the aerodynamics of it. So probably not that good of an idea. I dunno. I feel like there is room for improvement somewhere though.

Re:Any ideas for improvements?

[blackiner]

To expand on this just a bit: I feel like the landing platform itself should somehow be helping the rocket land. It is like when you toss a ball up and catch it. If you just leave your hand at a certain height, the ball smacks into it real hard. But if you move your hand down with the ball as you catch it you provide a much smoother deceleration and it lands more softly. Obviously finding equipment and mechanisms that could assist a rocket will be hard though (massive ass weight and extremely hot fuel).

Re:Any ideas for improvements?

[Bruce Perens]

Speed and distance are the big problems. Rockets can do both. Things that are tied to ships or the ground have trouble keeping up with the rocket.

Re:Any ideas for improvements?

[Anonymous Coward]

Maybe just make the important bits waterproof and drop the thing in the water under parachute? Seems a lot softer and more proven technology...almost guaranteed success.

Re:Any ideas for improvements?

[Immerman]

I suppose magnetic/sticky harpoons might help (you don't want to damage it after all), sort of like guy-lines on an airship. I don't know about a shock-absorbing landing pad (your catcher's mitt) though - in the moments before landing the backwash would be subjecting it to forces exceeding those of the weight of the rocket itself - probably at least as difficult to tune any "give" to occur at the proper moment as i is to land the sucker in the first place.

Plus, as others have mentioned, Musk seems to have his eye on Mars. Landing a colony ship can probably only be done by rocket, and there won't be any special landing pads on Mars. Plus, for more terrestrial concerns, if he can master landing on a simple barge, he can land pretty much anywhere, which dramatically improves the value of his rocket design onte international market: any bit of flat, stable land is a potential cheap spaceport that his rockets can service.

Re:Any ideas for improvements?

[Anonymous Coward]

Actually I was wondering that they didn't use a parachute as a air brake to decelerate it to 2/3 or lower of the descent speed, it has the advantage of maintaining it upright and don't interfere much with the maneuverability. Also for this scheme the parachute can be actually quite small. Basically the idea is to ad a small parachute to maintaining it upright while using the motor as main descent control system.

Re:Any ideas for improvements?

[ColdWetDog]

Rotating a thousand ton barge at 60 rpm would be interesting.

Re:Any ideas for improvements?

[ColdWetDog]

A parachute would really limit your maneuvering ability. You've essentially made yourself a very long pendulum. Then you have to get rid of it. And high speed parachutes aren't exactly simple tech either.

Re:Any ideas for improvements?

[Anonymous Coward]

But have you ever tried catching a *rocket ball* without moving your hand? Totally different experience.

Re:Any ideas for improvements?

[Bruce Perens]

The final approach is at 250 m/s. If I have this right, they'd be going about that fast if they started falling from zero velocity at 3 KM, ignoring air resistance. So, whatever parachute you use has to get you much lower and slower than that, and so precisely positioned above the barge that you can do the rest on the rocket.

Now, ULA plans to revive the Rogalo Wing from Gemini and combine it with the mid-air retreival from Corona, so this might not be completely absurd.

Re:Any ideas for improvements?

[cjameshuff]

More than that: since the minimum thrust is greater than the vehicle weight, slowing it down would actually mean a shorter landing burn. The faster the stage is falling, the higher it starts the landing burn and the more time it has to correct its descent.

Note that they don't fully extend the legs until just before landing...this might be to reduce drag and ensure it has enough descent speed to make the landing.

Re:Any ideas for improvements?

[Anonymous Coward]

I don't know about a shock-absorbing landing pad (your catcher's mitt) though

It occurs to me you could do that with a hard pad on a structure or surface that has some give, like a shock absorber. Or maybe something pneumatic that could sink a bit from the shock and raise back afterwards, like ... a boat ... floating on some sort of ... liquid ...

Re:Any ideas for improvements?

[SuricouRaven]

How about using a non-high-speed parachute? Rocket it down to low velocity a kilometer up, deploy chute. You won't be able to aim very precisely, but you just need a big, flat expanse of land, and those are easily found.

Re:Any ideas for improvements?

[Immerman]

I'm fairly certain such passive systems only make the problem worse - as I said the backwash will be subjecting the platform to forces exceeding the weight of the rocket: the rocket is decelerating (Fa greater than Fg), and at close range the wake is all hitting the landing platform, transferring the same force as to the rocket. So the platform is already at "maximum give" before touchdown occurs, and the instant the rocket is shut off the platform will rebound, increasing the impact forces if touchdown hasn't quite occurred yet, or at best "bouncing" the landed rocket.

Re:Longer Video Shows How Incredibly Far...

[Anonymous Coward]

You have never encountered an "Elon Musk fanboy".

Re:Longer Video Shows How Incredibly Far...

[ColdWetDog]

You're just jealous 'cuz you're mom won't by you a quadcopter.

Speaking about getting a life ....

why dont they spin it? and land it in a silo?

[sberge]

A bit of rotation should help to keep the thing upright. The gases being pushed into the silo will be forced towards the walls on their way back out and help center the rocket as it enters the silo. A funnel-shaped silo is easy to hit and provides a soft cushion as the pressure of the backscattered gas increases as the rocket descends into it. Finally, a rotating platform at the bottom needs to be synchronized to the rocket's own rotation. Good Thing I don't have the billions it would take to see my brilliant ideas crash and burn.

Re:why dont they spin it? and land it in a silo?

[diamondmagic]

I like this idea. I'm thinking, though, once you've already programmed a rocket to be able to land in a particular place, getting it to land without tipping over isn't that much extra work, compared to the cost of building out an entire silo with spinning platform.

And I guess they wanted to stick with an ocean landing for some reason.

Re:why dont they spin it? and land it in a silo?

[Caviller]

Yes, that would help stabilize the rocket...but unfortunately...it would cause centrifugal forces on the fuel and cause the fuel to pull away from the fuel pickup at the bottom center of the tank. They are already quite low on fuel at this point so it won't take much spin to kill the engines.

Re:why dont they spin it? and land it in a silo?

[Anonymous Coward]

What would you put on the landing gear? Wheels?

Re:why dont they spin it? and land it in a silo?

[towermac]

It doesn't have to be a big heavy concrete silo. Picture something like a big playland plastic thing, like a too deep skateboard well at the park. Maybe 15 degree angles; the rocket at rest can lean against it without undue mechanical stresses. It covers the whole barge, and it's tall, but light; so it is still a viable ship.

But a huge target, that can't be missed, and you won't tip over.

Downside; can't relaunch from the barge now. Have to just ferry it back to the Cape.

Seems like a pretty good idea though.

Re:why dont they spin it? and land it in a silo?

[tlambert]

I suggested the funnel landing previously. You really wouldn't want to spin it.

I'm pretty sure the problem at that point would be thermal damage to the vehicle. You could perhaps do this on land, and of course you'd probably want fresh water as a coolant and noise suppression, same as the shuttle used to suppress acoustic energy reflected from the pad to prevent it damaging the shuttle. The shuttle usually went through about 1/3 million gallons of water per launch. Of course with the rocket exhause itself, that produced a lot of HCl, and you probably wouldn't want to sit your engines in that, any more than in the ocean.

It's probably be better if they just had flip-down feet to widen the base of the cone ( the falling over was always a result of the center of mass being outside the interior circle of the cone, as described by the landing base -- or it wouldn't have fallen.

The other suggestion would be a big-ass electromagnet to lock the base components into place as it currently exists, but getting power out to the platform for that might be problematic. Maybe they could use a bunch of Tesla batteries, since it's only have to hold until you could mechanically latch the landing legs down.

Re:why dont they spin it? and land it in a silo?

[Anonymous Coward]

If it's spinning, how will they use the RCS? If you watch the video, after it touches down an RCS thruster starts going full blast to try to get it vertical. If the thing was spinning, there's no way they could attempt that, since the thrusters would be spinning around as well.

And the whole "specialized structure" thing: it'd cost a lot, it'd be damaged by failed landings and need expensive repairs, and it wouldn't help much when it comes to eventually landing a Dragon-derivative on Mars (or whatever their current long-term plan is).

Re:why dont they spin it? and land it in a silo?

[sberge]

You could have a silo in the ocean too. A heavy one made out of concrete would stay mostly submerged.

Re:why dont they spin it? and land it in a silo?

[sberge]

If you offset the fuel pickup from the center, the spinning might possibly increase fuel pressure.

Re:why dont they spin it? and land it in a silo?

[sberge]

The spinning would have to be slow enough that you could either gimbal the nozzle or modulate the thrust at the same frequency as the spinning. I don't know which of the two controls is faster. I don't know about the economics of this. It's probably not economical.

Re:why dont they spin it? and land it in a silo?

[sberge]

That would be the reason for the spinning platform at the bottom of the silo ;-). But maybe it's better to spin down the rocket as it descends so that it stops just as it lands.

Some Additional Details

[DanielBigham]

- Musk has commented that the issue has been diagnosed as some stiction that was causing a lag between computer commands getting carried out. They believe the issue will be straightforward to fix. - Musk's claim is that the barge didn't sustain any serious damage.

Re:Some Additional Details

[Anonymous Coward]

this is the same musk that claims autonomous vehicles and general artificial intelligence will happen. they can't even get a rocket to land safely on a barge.

I've been studying AI for over a decade now and all of these people sounding alarms about imminent strong AI are talking out of their butts. In MIT review it has a decent article debunking the non-sense from people who know very little about AI. http://www.technologyreview.co...

obvious damage

[enzo1]

"Droneship is fine. No hull breach and repairs are minor. Impact overpressure is closer to a fast fire than an explosion."â"Elon Musk

Re:obvious damage

[ColdWetDog]

Only a billionaire can say that damage from an exploding rocket booster landing on top of something is 'minor'.

Re:obvious damage

[Harlequin80]

It's a fireblast though rather than an explosion. When the rocket falls over the tank is ruptured and the fuel vapours pour out and are ignited. This then results in a flash fire and a pretty impressive fireball. There is nothing of any real strength containing the blast pressure so most of the energy is dissipated harmlessly into the air.

You would probably find the surface of the vessel is something like thin concrete over a steel plate. The concrete will likely only have minor damage that can simply be patched. If the hull hasn't been breached and the engines undamaged it is good to go.

Re:obvious damage

[complete loony]

When the rocket booster has already done it's primary job and would otherwise be lost anyway. The only real loss is the damage to the barge.

Re:obvious damage

[gweihir]

Actually, any competent engineer can. You mistake "Hollywood" explosions for ones that do actual damage.

Funnel

[sexconker]

Build a large funnel, or infundibulum, on the barge.
All you have to do is hit the wide top of the funnel at a non-clusterfuck angle.
Then you let the structure of the funnel contain and guide the rocket as it continues on down.
If you fuck up badly, you won't lose everything. And if you do very well for 98% of the landing but tip toward the end, damage from impacting the walls of the vertically will be incidental and minor as the rocket is still thrusting to lower its velocity (and thus the force of the impact).
And if you do it successfully the funnel isn't touched.

Alternatively, do the same but instead of a solid funnel, use closing arms so you can actively catch and assist the rocket if need be, or drop the arms and let the rocket fall into the ocean if you have to abort.

You could also put 8 electromagnets in a circle with the target in the center. If the rocket leans north west, increase power to the southeast magnet and decrease power to the northwest magnet.

Another idea would be to have a guy with a long stick on the barge ready to nudge it just a bit if it starts to tip.

Re:Funnel

[Applehu Akbar]

Simpler tahn closing arms would be closable netting in the upper half of the funnel. As has been pointed out above, a booster is strong in only one axis, so you don't want any points of concentrated stress in any other direction.

Re:Funnel

[Harlequin80]

I believe the aim though is to require no special landing equipment. Given these tests are essentially covered in the launch price for the ISS SpaceX are likely to keep trying unassisted landings until they either succeed or say it is outside their capability.

Re:Funnel

[sexconker]

The problems with netting are:
You need to support it fairly evenly with some sort of structure anyway.
It has to be able to withstand those temperatures.
It has to be decently strong, which means a fairly dense netting pattern, which means more weight to the netting itself that you must support with some structure.

Lining a funnel or the action areas of a set of arms with some sort of padding would be as effective as netting. A full-on net would get you more leeway in actually catching the rocket, but they've almost got the vertical landing down anyway.

Re:Funnel

[sexconker]

No special landing equipment.
Except a platform able to withstand the thruster. On the open ocean. With people monitoring the landing. And crews ready to rush in once landing completes (successfully or not).

It may one day be as routine as parallel parking, but basic equipment to help corral the rocket and prevent catastrophic failure is a no brainer. Saying you don't want any special equipment because you want to be able to do this without it is like saying self-driving cars shouldn't have air bags and seat belts because you want them to avoid crashing in the first place

Re:Funnel

[Harlequin80]

Except what if you want to land where there are no crews.

The barge is there because they are currently not allowed to land on the ground. The people monitoring it are there because this is still experimental. I would expect that a funnel would create all kinds of hard to predict blowback from the rocket itself and present its own engineering challenges.

If you consider Musk's stated goal of landing on Mars then landing without anything other than a flat patch of ground becomes necessary.

Please cite original sources!

[Anonymous Coward]

The article is not the original source of the video. It's just an embedded YouTube video. The actual YouTube page for the Falcon 9 first stage is landing is

http://youtube.com/watch?v=BhMSzC1crr0

Not 'close'

[YrWrstNtmr]

You need three conditions:
1. Hit the target location
2. Minimal vertical velocity
3. Vertical orientation.

They met #1. It was coming in tilted (for varying amounts of tilt), and way too fast.

Re:Not 'close'

[Bruce Perens]

They planned the fast approach. Consider that the main control surfaces are the grid fins. They don't work at slow speeds. It's all about landing with the minimum use of weight (thus fuel).

Re:Not 'close'

[Pope Hagbard]

They got #2 as well, it was just that #3 wasn't quite there.

Re:Not 'close'

[bledri]

You need three conditions: 1. Hit the target location 2. Minimal vertical velocity 3. Vertical orientation. They met #1. It was coming in tilted (for varying amounts of tilt), and way too fast.

It did not come down way too fast. It just overcorrected at the end. It's called a hover slam (unfortunately) and it's all about using the minimum amount of fuel and dealing with the reality that even a single engine throttled down to 40% has a T/W > 1. They are extremely close to successful landing a first stage booster used to deliver a payload to orbit. Only experimental rockets have done this before (landers don't count, they are tiny in comparison. Amazing feets but a different kettle of fish.) This is a big deal. This is going to make spaceflight cheaper.

Re:Not 'close'

[Anonymous Coward]

It's supposed to come in that fast, it's a hoverslam. Their only problem was actuation latency causing a small oscillation in the horizontal. Next time it will come in just as fast, and land perfectly vertical. If it looks fast for you - it's because no human could ever pilot it, but the computer can.

Re:Not 'close'

[Anonymous Coward]

They got #1, #2 and almost #3. They just had a bit too much momentum and lost #3 at the end.

Re:Not 'close'

[Anonymous Coward]

Hum... I would think that the side motion was done to avoid burning through the pad. I doubt they would approach the thing in direct line...

Re:Not 'close'

[michelcolman]

Actually, I think #3 was there, but #4, zero rotation speed (the derivative of #3) was not. It touched down pretty much vertically, but still rotating. Just before touchdown, the rocket seemed to become dynamically unstable because the rocket exhausts moved with a bit of lag. They keep firing in the same direction while the rocket passes throug vertical and in the other direction, and only then start correcting in the other direction. They should actually have changed direction before the rocket was vertical so the rocket would not just return to vertical, but stop rotating as it reaches vertical.

Re:Not 'close'

[Anonymous Coward]

1 right place (x,y,z)
2 near zero velicity (dx,dy,dz)
3 right orientation (angX, angY, maybe angZ)
4 near zero rotation speed (dangX, dangY, dangZ)

which of these 12 did the actually have under cereful control?

It looks like the ang's, dang's, and dizzy's had some room for improvement?

Re:Not 'close'

[Anonymous Coward]

Condition 4: ability to stay put once they get there.

at 1080P and single stepping the frames,

They got pretty close to all 3 at 8 seconds, then it fell apart at 9 seconds.

It looks like the bird did a sufficient job if the boat had just reached out and grabbed it.

Truly awesome that they got this far given what they are working with.

Re:Not 'close'

[cbhacking]

Yep, Also, while it's (obviously) true that they didn't stick the landing vertically, they weren't far off. For that RCS thruster at the top to have held the rocket upright as long as it did, the stage must have been *barely* past its center of gravity. The thruster (and remember, these are simply compressed gas thrusters intended to impart a quick nudge; they aren't very powerful) fired for about 4 seconds continuously, which probably exhausted its propellant, but in that time the stage didn't visibly lean over any further!

A *tiny* bit closer to vertical, and the thruster would have been able to correct. A bit beyond that, and the thruster wouldn't even have been needed; if the center of gravity is inside the footprint of the landing gear then the stage would self-right as long as the legs didn't crumple (and, contrary to what the OP says, they actually had excellent control of vertical speed). Alternatively, a slightly more powerful thruster would have been able to correct the angle (I wonder how much it would cost - mostly in weight, I assume - to switch to hydrazine rockets like the Dragon's "Draco" RCS thrusters).

Compare to the first landing attempt, where the uncontrolled guidance fins meant the rocket was coming in at such a sharp angle that it basically skipped off the barge like a rock on a lake. This one was *much* closer to upright, and while it did fall over, the whole thing (achieved zero vertical velocity resting on its landing gear for a moment. That's damn impressive, and it was a very close thing.

Re:Not 'close'

[Bruce Perens]

Here is the barge on the way to port, possibly with debris onboard. Here's a video of the landing shot from the barge itself. And I am waiting to see the barge from the Carnival Fascination webcam.

Re:Not 'close'

[tibit]

Sorry, they met #1 and #2. The vertical velocity at contact with the barge was as close to zero as anyone could wish for. That's the only one that matters, given that it's a system with thrust-to-weight ratio > 1. The on-contact vertical velocity was lower than the vertical velocity on 95% of commercial transport airplane landings when the wheels touch the pavement. You'd be really surprised at how good this landing otherwise was, given the just-about-oscillating thrust controller issue that they had.

Re:Not 'close'

[tibit]

That thruster did the correction, but there was literally no leg for the thing to stand on. One or two of the legs have collapsed on landing. A bigger thruster would simply keep it upright until the nitrogen ran out (here it didn't), it'd still fall and go kaboom.

Actual solutions

[Anonymous Coward]

If they had a 16 foot fence around the barge at least they would have recovered the critical motor pieces. The tank contents is what "exploded".

I am fortunate to have had three of the lunar lander entrants do practice runs at our test site. I have seen liftoffs, translations, hovers and a wide variety of "landings". The key seems to be having a clear hover moment before committing to landing. It may also need an anti-slosh system at the bottom of the tanks. Especially the top tank.

They should rename the barge again, "Hit me, I need the money!"

New product

[SuperKendall]

Musk's claim is that the barge didn't sustain any serious damage.

Screw self-landing boosters. What I want is a house made out of whatever the barge is made of, easily shrugging off what are essentially two direct rocket hits complete with massive explosion.

Re:New product

[Bruce Perens]

Just look for one that says "Fallout Shelter"

Re:New product

[AJWM]

This is actually a little-known third experiment that's part of the launches. They're perfecting the material to make Elon Musk's super-villain lair out of.

Re:New product

[petermgreen]

The rockets were basically empty at the time of landing. I would think a booster full of fumes would give something similar to a "hollywood explosion". Big fireball but relatively little bang.

Re:New product

[gweihir]

There was zero actual explosives in there and there was a showy, but mostly harmless fuel combustion which does not qualify as "explosion". Seriously, they do this in the movies because it looks impressive and at the same time does very little actual damage.

Re:New product

[Bruce Perens]

See this tweet.

Re:New product

[Bruce Perens]

It looked to me that the barge was structurally undamaged but that some heavy equipment on the deck was forcibly ejected. It's clear to see in the HD version. Those 1000 HP thrusters are expensive, and it looked to me like one of them going overboard. But I suspect they were prepared to lose more than one vessel in testing this.

And I bet there was a range safety self-destruct charge onboard. F9R blew itself up with one. But it was probably so safe that it didn't go off.

Re:New product

[Enigma2175]

There was zero actual explosives in there and there was a showy, but mostly harmless fuel combustion which does not qualify as "explosion". Seriously, they do this in the movies because it looks impressive and at the same time does very little actual damage.

The Falcon 9 does have a Flight Termination System to destroy it if it goes off course, so there may still be explosives on the rocket when it is landing.

Re:New product

[cjameshuff]

I'm pretty sure the large object that went overboard was the first stage's engine/octaweb section.

The range safety charges should be almost impossible to detonate by fire or impact, and there was an announcement that they were "safed" during descent, but I wonder if there might be a contingency for setting them off in situations like this...not to destroy the vehicle, but to make sure there aren't any surviving bits of high explosive strewn around.

Re:New product

[Bruce Perens]

A video from the barge is now online here. If you step through the final frames, you can see that the camera mount ends up knocked over and pointing at the ocean, but the lens and its cover are unbroken and all we see flying appear to be small debris. So not a really high-pressure event.

Re:New product

[Anonymous Coward]

What you call a massive explosion is just a conflagration of some rocket fuel fumes. There's barely enough overpressure to throw some bits of sheet metal off the barge. Heavy, solid parts like the engines probably weren't moved by the fireball to any significant degree because the gases were free to expand into open air. So there wasn't anything dense enough moving fast enough to hurt the surface all that much. If it's constructed like aircraft carrier flight decks, it's made of steel with a coating of high-friction "paint" on it (textured like coarse sandpaper). A fireball like this would only damage the coating.

Re:New product

[gweihir]

You can only get a high-pressure event if you finely distribute the fuel in the air before igniting it. (Traditional fuel-air explosives have one charge for distributing the fuel and one for detonating it. Other designs exist.) Everything else is just a flame.

Re:New product

[gweihir]

And where did you miss the part that I was obviously talking about what was going off? Also, explosives used for this task do not go off when exposed to flame, they just burn harmlessly.

Re:New product

[tibit]

The explosions aren't massive. The amount of propellants involved is rather small. As far as rockets-going-boom, this is nothing. Also remember that rockets used in warfare often carry conventional warheads that have an order of magnitude more stored energy than the propellants do. So even if the entire thermal energy from the propellants was converted to kinetic energy of the rocket (a thermodynamical impossibility) and delivered in the entirety to the target, it'd still be pointless.

TL;DR: That barge could withstand a fire from a large LOX and LH2 spill, but it wouldn't withstand a single hit from a 50x lighter rocket designed to actually damage things by going boom.

Re:New product

[tibit]

These explosives are actually hard to set off. If I gave you some, and you had nothing explosive in your house, you actually couldn't set them off.

Re:Close? Really?

[beelsebob]

It landed just a degree or two short of the RCS thruster being able to right it. It wouldn't have been perfect, but it would have been a landing and recovery.

It was only one iceberg...

[Anonymous Coward]

It was only one iceberg that sank the Titanic. It still sank. -- Jeremy Clarkson

both, like a dragster. Chute would help stabilize

[raymorris]

It occurs to me that a small chute at the top would help keep it vertical for a landing, as well as slowing it down a bit. Like the tail of a kite or dart, even a small amount of drag really would help keep it straight.

Top fuel dragsters show that a chute can be combined with other types of braking effectively.

Hollywood explosions..

[mutherhacker]

... ain't got shit on SpaceX. I always thought that on Hollywood they overdo it with explosions but if I saw this in a Hollywood movie I'd be like "Come on...Really? would the real rocket explode that violently? ". I stand corrected.

legs too late?

[short]

Aren't the legs stretched out needlessly late? It does not seem to me they were already fully deployed and fixed during the final touch down.

Re:legs too late?

[Bruce Perens]

In the F9R test videos they catch some of the backscatter from the engine and seem to catch fire. Maybe they were trying to avoid that. They are very light carbon composite. Or perhaps they mess up the airstream for precision navigation, or they don't like the 250 m/s wind.

Re:both, like a dragster. Chute would help stabili

[Anonymous Coward]

It has grid fins at the top of the first stage for this. Believe it or not, they understand the idea of a center of pressure vs a center of mass.

Two ships and a net?

[Anonymous Coward]

Just two ships with a net between them. Let the lander hit the water (slowly) and do recovery that way (unless water will do damage). Wet and intact might be better than "on solid barge but blown to bits". I guess it all depends on the structural integrity of the stage (water is dense and might bend/twist) also there is heat/thermal shock to consider.

Does not look close at all...

[gweihir]

The rocket is swaying severely. This could not have succeeded at all. The only thing keeping the rocket upright for a while seems to be some top-mounted side-thruster. Having the rocket hit the barge at zero speed is _not_ enough, it also has to be in a stable position while doing so. I do not see that here at all.

They will have to do some major revisions to the software, including aborting the attempt to protect the barge when the rocket comes in this badly unstable.

Re:Does not look close at all...

[tibit]

The rocket didn't come in unstable, it was pretty much following the barge avoidance reversion trajectory. The thrust output was unstable, but given that the trajectory was quite spectacular. Remember that for the rocket to end where you want it to be, you need to precompensate by turning the whole body sideways, so that you'll have countertorque for sideways thrust when retargeting to the barge. That seemingly large and "weird" excursion was completely normal. Try it out in KSP if you must, and remember never to throttle down beyond TWR=2.2.

Re:incredibly close to target is far from success

[Anonymous Coward]

It is mighty difficult to hover when your engine minimum thrust is higher than your weight. Your vision of the landing is physically impossible.

Re:incredibly close to target is far from success

[Smidge204]

The mission was complete; the cargo was delivered to the intended orbit with no difficulties.

They just didn't get the bonus points for a successful experiment in first stage recovery. Once first stage recovery becomes routine, then you can consider it part of the operation - but never part of the mission. They are contracted and paid to deliver the payload to orbit, not recover the first stage.
=Smidge=

Incredibly close?

[Anonymous Coward]

Bruce Perens. Shoulda known he'd be a Elon MUsk fanboi. For anyone with experience with control systems, the input corrections on its final approach were massive, as were oscillations on the vehicle. It hit the barge desk, but it wasn't close to a successful landing. My guess is the fuel budget is so tight, the rapid landing sequence is necessary. I enjoyed finally seeing the explosion.

Re:Incredibly close?

[tibit]

For anyone with actual experience with control systems, this looked pretty much spot on. The only oscillations just above critical damping were in the throttle value, and SpaceX acknowledged as much. I don't know what you really know, but it doesn't seem like much.

Re:incredibly close to target is far from success

[jo_ham]

So, with your expert knowledge of the situation, how do you propose to make the thing hover when the TWR is 1.8 at minimum thrust?

Just curious.

The vertical speed was also planned. How do you propose to control the rocket's attitude with the fins when it's moving very slowly, since they rely on airspeed to function?

Let me guess, you just assumed that "hover slowly and touch down like a helicopter" was the desired descent profile without looking anything up?

Other word for failure

[Anonymous Coward]

"incredibly close to success" is a synonym of failure. (but yes, it seems close. Hope they will make it. Lower price for launch means more satellites/probe demands, so more work for me)

Re:incredibly close to target is far from success

[Anonymous Coward]

You can PWM the engine. But I don't know what's their minimum opening cycle, and it will probably damage too much the engine/structure, defeating the purpose of landing to reuse...

"Close" isn't good enough...

[Anonymous Coward]

"Close" only counts for hand grenades and horseshoes.

Seems to me the landing barge needs to help more

[st0rmshad0w]

So why wasn't the landing barge designed with several pairs of masts or arms that would remain extended outward from the platform edges at a low angle, say 30 degrees, and then, once the rocket reaches a certain altitude, quickly raise on hinges to lean into the center and wrap the rocket in strong nets of webbing strung between them?

Wouldn't that go a long way toward safely keeping the vehicle in the vertical without being overly dependent on the small maneuvering thrusters?

This whole article is a blowjob for Elon Musk

[HBI]

We already want to change every law for this billionaire fucker. Now he gets to not fail when he fails.

Elon wants to go to Mars...

[Anonymous Coward]

...so we should imagine a rocket landing on Mars and making use of rocket fuel we flew out to Mars to refuel that rocket to send it back into Mars orbit before it returns to Earth.

At the risk of sounding like a fanboi, Elon dreams big. But it's all in bite-sized pieces, as genius always is. Only later do you look at it and say, "Well, it was quite obvious...". But at the time, it wasn't so obvious until someone came along and realized their vision.

Re:incredibly close to target is far from success

[Bruce Perens]

It's very tempting to think this should work like an airplane. Lots of people wrote that it was "too hot", etc. But it isn't an airplane. The plan was really to approach at 1/4 Kilometer Per Second, then brake at the very last second.

Obviously Crew Dragon, which carries people, will approach differently. But it's a lot lighter.

Elon's King of Swamp Castle

[Thud457]

When I first came here, this was all swamp.
Everyone said I was daft to land a rocket on a barge, but I built in all the same, just to show them.
It sank into the ocean.
So I built a second one.
And that one sank into the ocean.
So I built a third.
That burned down, fell over, and then sank into the ocean.
But the fourth one stayed up. And that's what you're going to get, Son, the strongest rocket in all of aerospace.

like CORONA, in The Road Warrior

[Thud457]

I was going to suggest a tower with a grapple just tall enough to hold the returning stage suspended.
But recent events have convinced me that snagging it out of the air with a giant auto-gyro and flying it back would be the most awesome solution.

Unfortunately, the nearest volcanoes are in the Caribbean. Although operating from St. Kitts might help with recruiting.

Re:like CORONA, in The Road Warrior

[lucien86]

Catching it with a big helicopter in mid-air might actually be a really good idea. It would take a huge helicopter though - and a huge net. :)

Maybe another idea might be some kind of last second purge system to flush the fuel out of the engines. Whatever, this near miss is still already a huge success.

Flame was not the only thing that was hot...

[EricTheO]

In the video, it looked like the booster came in too hot(fast) on final approach. This necessitated more severe corrections just prior to touchdown.