I think the coolest part about that video is how the aerodynamic guidance is very noticeable. Look at the significant angle between the rocket body and the velocity vector (as indicated by the contrail) and you can see how the rocket isn't just falling ballistically but is "flying" towards its intended touchdown point.
Curious how they are tracking the rocket's descent to catch it on camera. It seems like they lose visual of the rocket and then are able to re-acquire it fairly quickly. Are they using some sort of infrared or other tracking beacon on the rocket that the camera can follow or using trajectory data to determine where the camera should be pointed?
Also, I'd love to know when these launches are going to happen further in advance. The Launch Manifest page just shows the next launch as happening in June[1]. Is there a better place to get more exact planned launch dates further in advance?
Probably Radar based tracking with the possible add-on of multiple visual angles being able to communicate with each other. If you are interested in some further reading, here are some interesting articles/videos:
Thanks for the links! I'll try to look at those when I have some time. On a quick glance, it looks like most of those systems are ground based and pretty large (which works great for launch footage), the video posted appears to be looking down on the landing pad at the end, which leads me to believe the camera is located on a helicopter.
Could a radar based camera be mounted on a helicopter, or could land-based tracking devices work in conjunction with a helicopter mounted camera to achieve the tracking?
With the tiny embedded viewer, I also lost track, but on full screen, I was able to (barely) keep track of the rocket the entire time. I also assume they have a pre-computed flight path that's pretty accurate, maybe GPS on the vehicle and undoubtedly some radar tracking. In short, I can't imagine them not knowing precisely where that rocket is the entire time.
Unless something goes wrong. Well, several things.
I was wondering the same thing. To me, the camera work looks very confident (smooth, even panning). Almost as if they knew exactly where the rocket was behind the clouds. Infrared sounds plausible. Or maybe they just have an high-res camera that captures a large portion of the sky. Some student would then post process the video by zooming in and panning accordingly.
Awesome new video of the second landing attempt. Glad they identified the issue, can't wait to watch the next attempt! Also looking forward to them trying this on land. I live in the Cape Canaveral area, so it'd be pretty cool to be able to track a rocket coming in for a landing.
Didn't know it was that soon- awesome! I do wonder how much PR is motivating these blog posts. I'm sure an incoming rocket (basically, a missile) would scare local citizens, so it won't hurt to get these positive tests and lessons learned out in front of people.
There has been a few mentions of it over the past month or so that have made it to reddit.com/r/spacex(Gwynne Shotwell was the first to mention it) for the JASON-3 mission in august the jist so far is the barge will be off the coast (but relatively close) and a decision for RTLS or Barge Landing will take place just before launch
It seems like they could replace the landing legs with a parachute, which they could catch in mid-air, as intended for the Genesis probe. Bringing the 1st stage nearly to a stop at some point mid-air should be way easier than landing it on the ground.
For that matter, could a balloon be used as a recovery method?
Another alternative: Land in an artificial pond. The lake could be filled with a liquid specifically formulated to be kind to the rocket parts. (Perhaps just reservoir quality water?) The design could again dispense with the landing gear. Perhaps sturdy heat-resistant ballutes could inflate to take the water-impact and hold the rockets up and out of the water?
EDIT: Wow, so much anger! For one thing, all the above schemes involve first coming to an almost-hover mid-air. Also, why is everyone imagining such an amazing tone from my post? Please produce a quote that justifies this assumption. (Hint, it's irrational. Just take my brainstorming at face value.)
It always amazes me how people see this video and immediately think "I need to redesign this for them" rather than "wow, that prototype is amazingly close to working beautifully".
It's as if a complex software system is being tested for the first time, a couple bugs are found, and onlookers think the right answer is to redesign everything from scratch.
I'm picturing footage of early NASA rocket failures if the internet had existed with real time feeds. NASA would have been portrayed as an agency that couldn't accomplish anything. Keep in mind that while SpaceX is building on the previous hard work of rocket scientists around the globe they are also pushing the boundaries of what's possible. Very impressive to watch.
I'm picturing footage of early NASA rocket failures if the internet had existed with real time feeds. NASA would have been portrayed as an agency that couldn't accomplish anything.
Such portrayals were present in old media, and this is way far off from anything I'm trying to express.
Not the same. In 1960s people couldn't send around links to videos of exploding NASA rockets.
Following will have vastly different effects
1) Seeing a video footage once a life time
2) Seeing a footage (even edited for added entertainment) over and over again whenever you want
I'm glad NASA wasn't trying to develop rockets in the age of Facebook and youtube.
I am also amazed that people seem to want to replace the SpaceX system where the rocket would land itself at an arbitrary location on land, ready for re-fuelling and re-use, automatically, with the existing launch engines and no extra technology required beyond somewhere flat - with a complex system of tracking radars to find the rocket, extra aerodynamic drag devices such as parachutes or balloons to slow it down, and a fleet of manned vehicles to retrieve it, using a dangerous mid air rendeszvous. It's pretty obvious that these systems have none of the useful properties of the SpaceX solution, so it's hard to understand why anyone would think they are superior....
Just relax, man. I'm not claiming that I have "the answer." Seeing the landing 1st stage just makes me wonder about all the other possible schemes. (Probably the same thing happened when the Wright Flyer was first seen.)
You would come across more like you "just wondered about possible schemes" rather than claiming to have the answer if you had said "would it be possible to replace the landing legs with a parachute?" instead of "it seems like they could replace the landing legs with a parachute."
The way you formulated yourself clearly sounded like second-guessing.
Hacker News is harsh to people suffering from the Dunning-Kruger effect, and doubly harsh for those critical of SpaceX.
That said, your first comment really comes across about as aloof as some non-technical guy saying "how long will it take to build Facebook? It's a website right?"
The atmosphere of Mars is not thick enough to support a parachute landing, so that's out. Balloons do not work well on Mars either; landing heavy payloads with balloons result in a lot of bouncing, which is bad for both equipment and for humans. There are also no artificial ponds on Mars. Additionally, water is not good for the rocket parts (salt water is terrible, but I think even fresh water is not good).
SpaceX is iterating on technology that will eventually be used to land millions of people on Mars. Before they get to that stage, they are working on perfecting rapid, full reusability of rockets to 1) achieve lower launch costs to win the market, and 2) have a ready-to-go system for landing people on Mars.
The primary design principles of the landing system must work on Mars as well as they do on Earth.
Parachutes are not a good idea because of how rockets are built. The parachute would apply too much stress on the structure of the rocket which would reduce its lifespan/potential for reuse.
While it's normal to try to think and find better alternative, let me remind you that going into space is not simple. I always have a strong feeling about people over-simplifying hard problems with clueless alternative.
I feel this shows a lack of understanding and respect to a field you clearly don't understand. Try to take this as a constructive criticism.
Parachutes are not a good idea because of how rockets are built. The parachute would apply too much stress to the structure of the rocket which would reduce its lifespan/potential for reuse.
Thank you. I was waiting for this response!
While it's normal to try to think and find better alternative, let me remind you that going into space is not simple. I always have a strong feeling about people over-simplifying hard problems with clueless alternative.
I know I'm a layman in this field. Nothing wrong with asking questions.
I feel this shows a lack of understanding and respect to a field you clearly don't understand.
I know I don't understand everything, but you should know better than to just engage in harsh silencing tactics against anyone who simply asks a question out of curiosity. If you see something in my post suggesting this is not the case, please post a quote. Then if you can't find anything that isn't subjective "tone" then please know that this was all in your imagination. Try to take this as a constructive criticism.
You mostly didn't ask questions, though. Of the three possibilities you gave, the only one you expressed in the form of a question was the balloon one.
I get it, you were just idly speculating and wondering about possibilities. But surely you see why people are reacting badly to it given that you did not just ask questions.
What kind of equipment do you propose to use for a mid-air recovery of a fragile rocket stage that's 200ft long and weighs a couple dozen tons?
What kind of liquid do you propose filling the pond with that won't destroy the rocket on impact? Hitting water at high speed is like ramming into concrete, and I don't think other liquids will be much better.
Your ideas sound like, "What SpaceX is doing is far too easy, let's come up with vastly more difficult approaches so they have a proper challenge!"
What kind of equipment do you propose to use for a mid-air recovery of a fragile rocket stage that's 200ft long and weighs a couple dozen tons?
A large rig suspended from beneath a large helicopter. If there's a parachute involved, then it supposedly has a mount point that can support the weight of the rocket under high g load. (And of course, that would add weight.)
What kind of liquid do you propose filling the pond with that won't destroy the rocket on impact? Hitting water at high speed is like ramming into concrete, and I don't think other liquids will be much better.
Points off for poor reading comprehension. The scheme is to replace the landing legs with inflatable ballutes. The 1st stage would still nearly land just above the water. The impact the ballutes would take would only be from a short distance.
Your ideas sound like, "What SpaceX is doing is far too easy, let's come up with vastly more difficult approaches so they have a proper challenge!"
Your comment sounds a tad bit sensitive and emotional? In reality, I'm just going through the intellectual exercise of seeing what would work, given the landing control getting no better than what we have now. I fully expect it to get better, however.
The largest production helicopter ever made, the Mi-26, has a maximum cargo capacity of 20 metric tons, which is already insufficient. Making things worse, there would be a tremendous shock when the catch is first made. I guess if you want to add "building a helicopter far larger than any so far yet built" to the list of SpaceX's activities....
The stage will fall over on impact with water, so you're still talking about a 200ft drop for the top of it, and it's still going to hit hard when this happens.
You don't appear to have any concept of the numbers involved here or the physical limitations, so I don't see how this could possibly be an intellectual exercise of seeing what would work. Saying "just catch it in mid-air with a helicopter," when no helicopter capable of doing this currently exists in the world, and building one that's suitable would be a massive undertaking, is intellectually lazy, not an intellectual exercise.
I don't see where I'm being sensitive or emotional. I'm just pointing out that your suggestions are, in fact, a whole lot worse than what's currently being done, and are therefore kind of ridiculous to just suggest off the cuff. Brainstorming is fun, but is it too much to ask to do a basic smell test on the ideas before you start talking about them as if they were realistic?
The largest production helicopter ever made, the Mi-26, has a maximum cargo capacity of 20 metric tons, which is already insufficient.
There are startups going for cheap access to space with a much smaller class of launchers. 20 metric tons might well be workable for such a scheme, and it would lighten the control requirement under rocket power, which is admittedly hard.
Also, there isn't even a requirement for the aircraft to be a helicopter, so much heavier 1st stage boosters could be handled. Helicopters would enable much lighter logistical requirements, however.
You don't appear to have any concept of the numbers involved here or the physical limitations, so I don't see how this could possibly be an intellectual exercise of seeing what would work. Saying "just catch it in mid-air with a helicopter," when no helicopter capable of doing this currently exists in the world, and building one that's suitable would be a massive undertaking, is intellectually lazy, not an intellectual exercise.
You know, you're applying an awful lot of assumptions to make the general scheme not work: a particular booster, a particular kind of aircraft,... etc. You can also think of that as informing the limitations of the particular scheme and delimiting the size and purpose of the launches it's suitable for.
I don't see where I'm being sensitive or emotional. I'm just pointing out that your suggestions are, in fact, a whole lot worse than what's currently being done, and are therefore kind of ridiculous to just suggest off the cuff. Brainstorming is fun, but is it too much to ask to do a basic smell test on the ideas before you start talking about them as if they were realistic?
Your "basic smell test" takes very particular assumptions that produce a ridiculous result. It's intellectually lazy on your part to just leave it there.
I'm applying an awful lot of assumptions? This is a thread discussing SpaceX's Falcon 9 reusability attempts. Your comment started with, "It seems like they could replace the landing legs with..." which makes it pretty clear that you're talking about this particular equipment, not the general concept of rockets. When I asked what would recover the thing, you said it could be done with a helicopter.
You cannot propose using a helicopter, then come after me for "making assumptions" for looking at how it would work with a helicopter!
Yes. Your assumption of the particular launch vehicle for one. Your assumption of a particular kind of aircraft for another.
When I asked what would recover the thing, you said it could be done with a helicopter.
It is clearly your assumption about "it" being the Falcon 9 1st stage. Also, the general principle of recovery by aircraft doesn't necessarily mean it has to be a helicopter.
No, your original comment was clearly about the Falcon 9, what with referring to "they" (i.e. SpaceX) and talking about a specific feature of that one rocket (i.e. the landing legs).
You talked about recovering the Falcon 9 with a helicopter. When I pointed out that this wouldn't work, you then say I shouldn't assume either Falcon 9 nor helicopter??!??!??!?!?!?
Specifically, which large helicopter has the required payload capacity?
Also, rockets are built to support compressive loads. The tensile loads imposed by a parachute slowing the stage down are completely different, so there's no guarantee you could even use a parachute without significant strengthening of the structure.
Specifically, which large helicopter has the required payload capacity?
One might have to build one, and the scheme may well only be suitable for smaller rockets. Also, the point isn't to slow down the 1st stage from max velocity. It's to replace the landing legs after a near-hover.
> One might have to start an entire aircraft development program, employing a hundred or so rotorcraft engineers, constructing test facilities along the way, that might take years and tens of millions to complete
> Bringing the 1st stage nearly to a stop at some point mid-air should be way easier than landing it on the ground.
One thing to note about Falcon 9 is that it's impossible for it to hover since its thrust-to-weight ratio is greater than 1 at the time of recovery, even at minimum throttle. So you have to do a suicide burn such that you zero out velocity at your target.
Someone already pointed out the problem with helicopters capacity compared to the Falcon 9 weight. The idea is not without merit, However. ULA is considering a similar approach for the Vulcan. The big difference is that they plan to eject the engine core and catch only the engines with a helicopter. See http://spaceflightnow.com/2015/04/13/ula-unveils-its-future-...
The downside of this plan is increased complexity. Falcon 9's approach requires fins, some landing struTs and software. ULAs approach requires a system to eject the core, a shield for the cores re-entry, a parachute, a helicopter, and impeccable timing. I'm guessing that catching a parachute with a helicopter is an incredibly difficult problem just by itself.
I would suggest that you watch the Vulcan development to see how your thoughts play out.
Regarding the pool, I would assume that the issue would be finding an inert liquid, putting it in the middle of the ocean stably and enough quantities for a landing, and then getting the rocket into the liquid without damaging it. You'd likely still have all the same problems as just landing the rocket -- you have to zero horizontal elicits or the thing tips over and impacts the liquid too hard, and you have to zero verticle velocity or your impact is too hard. Ultimately you might as well just land the damn thing, which is also more versatile.
Please produce a quote that justifies this assumption.
How about "It seems like they could replace the landing legs with a parachute ..."? Or "bringing the 1st stage nearly to a stop at some point mid-air should be way easier than landing it"?
Rather than continuing to assert that there is absolutely nothing about your post that implies you know better, you really should drop the ego and dispassionately read your own posts from an outsider's perspective. Understanding how you come across to others will be helpful when formulating yourself in the future.
That's actually the approach being used with the ULA Vulcan to recover the first stage: https://en.wikipedia.org/wiki/Vulcan_(rocket). I think it is an easier method with less flexibility than self-landing. As typical SpaceX is shooting for the stars.
This (the whole chain of comments stemming from this parent in parent) is as ridiculous as it is entertaining. A lay person asks a question but uses phrasing that could be construed as declarative at first glance and suddenly everyone on HN is a rocket scientist, a SpaceX PR spokesperson, a SpaceX fanboy, or all three. Adorable.
The true irony is that all of the people responding with righteous indignation ( typically a good sign that the speaker has lost their objectivity) are 'defending' SpaceX from the violent attack from a person that is just Wonder if there may be another way to do something. Priceless.
And just to be crass, there are many fucking ways to skin a cat, assuming you get to build the cat, the skin, and the implement(s) to be used.
I'm pretty sure the wizards at SpaceX are ok with people thinking stuff and writing it online. All of you apologists may be better off sitting in quiet contemplation of the place the Crowd takes in literally every story of innovation.
It's kind of funny because "catch it with a helicopter" is ULA's plan, with the caveat that they're only catching the engines and not the whole rocket.
One of the design goals is the ability to land on Mars, and none of these ideas satisfy that constraint. Powered landing also has advantages in control, ease of recovery, etc.
Just landing on a larger target than their barge would be easier. The April landing probably would have succeeded if they hadn't had to make a big horizontal move just before landing to get into position over the barge. Once they get the barge landing working once, they plan to switch to landings on a landing pad, which can be made much bigger than the barge.
The horizontal correction at the end was part of the plan. I suspect the purpose is to come in at an angle so that the direction of the final correction is approximately known, and they can correct only by varying the amount of thrust, not the direction of the thrust (which is presumably slower).
Vs. coming in right on target - you'll still have small corrections to make at the end, but you have to gimbal the thruster in a random direction.
They come in at an angle so that the rocket is not over the barge for long, that way if it fails higher up it doesn't fall onto the barge with a lot of kinetic energy.
Once you realize that the end goal is to be able to fly back rockets in places with a minimum of infrastructure it makes sense. Heck we landed on the moon the same way.
Also, rockets are deceptively fragile so most variations of the 'catch and drop' type recovery won't be good for the rocket.
Something suspended from underneath a large cargo helicopter, or a large cargo helicopter specially constructed for the purpose. As in the Genesis probe, the device would be catching the parachute.
Mass of genesis probe: 494kg according to Wikipedia.
Mass of an empty F9 stage: not public, but http://www.spaceflight101.com/falcon-9-v11.html estimated it at 23,000-26,000 kg. That's almost twice the 14,000 kg payload of the heaviest helicopter I could find, the CH-53E. Do you really think designing and building a new helicopter with twice the payload capacity of anything built will be a more cost-effective plan?
That's incredibly light! It couldn't be done with an existing helicopter, but it could be done with an airplane.
Do you really think designing and building a new helicopter with twice the payload capacity of anything built will be a more cost-effective plan?
Specifically for the Falcon 9, no. For the record, I think SpaceX knows what it's doing, and if they've concluded landing the 1st stage is the best way to go, it probably is. But it's fun to think about spacecraft recovery and huge machines. (Unless you're being piled on by tons of emotional HN commenters who imagine you holding strawman positions and speaking in imagined tones.)
I wonder if you could use an air "force field" to capture the rocket. e.g. a circular arrangement of air turbines, dynamically adjusted to smoothly capture and land the rocket. Has something like this been tried?
I'm not sure if that's ever been tried, but it certainly won't work for SpaceX. The landing system has to be able to work on any solid, flat surface in the solar system, such as the surface of Mars. There are no force fields on Mars.
Edit: Every thread on HN or reddit usually has people suggesting other methods of landing rockets that require huge, specialized infrastructure. The primary goal here is to land on Mars without large or complex infrastructure. So even if you think you can devise a cheaper or 'better' system, if it doesn't work on Mars, it doesn't work at all.
The amount of CFM of air needed to lift a rocket booster would be immense. Back of the envelope math suggests somewhere around a football stadiums worth of air per second.
It seems to me that they'll have to use the rocket engine to slow the rocket down regardless of how it is recovered. Once that's being done it is just a matter of increasing precision in order to set it down with no additional complexity.
That said, there is complexity in having higher performance valves and such to get a fast system response, but still that's improved performance of existing equipment as opposed to adding more stuff.
