Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Rocket Lab - Reusability Plans for Electron Rocket via Helicopter Catching (rocketlabusa.com)
88 points by kevinsundar on Aug 6, 2019 | hide | past | favorite | 51 comments


Good on RocketLab for finally doing reuse. It's such an obvious path for improving flightrate and reducing cost for a high flightrate rocket that I'm super surprised that it has been mostly poo-pooed by the smallsat launcher startups (even though they all seem to be shooting for high flight rate). There are also a bunch of different recovery options available to smaller rockets that aren't available to large rockets. (And Rocketlab has shown an ability to execute and enter operational service that makes me fairly confident they'll succeed.)

By the way, this is nearly identical (minus the helicopter) to the method that John Carmack's Armadillo Aerospace (now Exos Aerospace https://exosaero.com/) used for recovering their Stig reusable sounding rocket back in 2011/2012. I'm a little sad that none of the media reports are mentioning that, as it's clearly a direct forerunner and Exos Aerospace is still doing flights.

Here is the ballute: https://www.universetoday.com/93281/armadillo-launches-a-sti...

Here is the guided parachute return: https://www.youtube.com/watch?v=UV7zL07Tof8

And here is an Exos Aerospace flight earlier this year: https://www.youtube.com/watch?v=_G9dJ3IQQVk


The key difference is velocity. Exos/Armadillo have never launched an orbital class vehicle, which comes down a lot faster and hotter.

I love Carmack and rooted for Armadillo. But ballutes are 50 year old tech (Gemini’s escape system had them.) Armadillo’s real innovations were in other areas and doing things faster and cheaper. Carmack would have been an interesting fit at SpaceX if he hadn’t gone to Oculus.


Of course it's harder and faster. But the recovery mode is the same, and that is notable.

Recovery of rockets that have been to space, even just suborbitally, is not trivial.


> Good on RocketLab for finally doing reuse.

"finally" seems a little harsh since their first commercial launch was less than a year ago. Seems like they've been working on it for some time and as mentioned in the presentation, RocketLab is usually pretty careful about what they announce.

They seem really smart with the strategic moves they've been making so I'm excited to see this play out.


Peter Beck is kind of famous for poo-pooing reusability in launch. That's what I'm referring to.

In part of the reuse presentation, he said, "this is a cap I'm about to eat because I've said publicly a few times the various things that rocket lab would never do so [i.e. reuse] unfortunately I find myself in the position of eating my hat." Here: https://youtu.be/joONWIGtcdY?t=501

Rocketlab has been able to execute very effectively, which is why I appreciate them. It had long bothered me that they didn't put their obviously impressive engineering prowess to bear on the reuse problem, and I'm very glad that they've come around.


They seem to prefer to quietly do stuff then announce it (their kick stage for example, the silly sparkly thing), given that I suspect they now have enough data to be really sure they can pull it off


Direct link to an animation of catching a rocket with a helicopter: https://youtu.be/joONWIGtcdY?t=858


Catching things in the air is hard...

But it's been done[1].

[1]: https://en.wikipedia.org/wiki/Fulton_surface-to-air_recovery...


Like he says in the video, catching is actually the easy bit.

Having the booster survive the re-entry heating and aerodynamic loads while only using passive methods to slow it is the hard bit.



More like these, given that the rocket will be falling rather than tethered:

https://en.m.wikipedia.org/wiki/Mid-air_retrieval


In case anyone's youtube is also not opening at the right spot, the animation is around 10:00.


Can't edit my top post for some reason but here's the updated link: https://youtu.be/joONWIGtcdY?t=583

Looks like the video got truncated after the livestream.


Same technique as NASA investigated for recapturing a Saturn V booster:

"[The helicopter] would be gigantic. The rotor diameter would be over 120 meters (400 feet). Its empty weight would be over 200,000 kilograms (450,000 pounds), with a useful load of nearly 250,000 kilograms (550,000 pounds), for a gross weight of a whopping 453,000 kilograms (1,000,000 pounds)"

http://www.thespacereview.com/article/3741/1


I've only watched the animation, not the whole video, but it seems like this is missing a crucial step. It seems to show the first stage magically start moving backwards, rather than continue to move forwards (coasting) and use thrusters to turn around for re-entry down-range. I'm sure the actual rocket experts will have it right, but is it me or the animator that's misinterpreted things?


It's just a combination of camera tracking and the acceleration of the second stage. The rocket follows a ballistic trajectory, there's no need for it to turn around. Or are you just thinking of orientation? The rocket will probably have RCS to do it's flip and re-enter with proper orientation.


Yes, orientation. The orientation doesn't change on screen, and yet it's going thrusters first to the right, which looks like where it's come from.


yeah I wondered about that too, I suspect it naturally arcs over and comes down heavy end first (a bit like a shuttlecock) with empty tanks the heavy end is the motors.


Yeah, I gotta get a new job.


Same. This got my envy meter charged up for a good week of feeling sad that I'm not working on things like this.


The failure modes on that look fun. I wonder how much they are planning on paying the pilot.


Should be falling relatively slowly at that point. Helicopter can be pretty distant until the rocket has slowed and stabilized, and then always remain above it. The chance of two things that small in 3D space hitting is pretty small.



This is great. I was hoping for a little more information in the presentation on the specifics, but still wonderful news. Best part was reading a Reddit forum before the announcement where someone was making a passionate argument that the presentation absolutely wouldn't be about reusability.


It's worth watching the animation video to the standup at the end where they explain a lot more about the issues


I watched the whole presentation. I was expecting Beck would go into the specifics about the problems and how they specifically are planning to solve them, kind of like Musk does. I’m sure we’ll get more details. I was just curious.


There's also a much longer version here

https://www.youtube.com/watch?v=joONWIGtcdY


I look forward to the video of Peter Beck eating his hat.


> The first phase will see Rocket Lab attempt to recover a full Electron first stage from the ocean downrange of Launch Complex 1 and have it shipped back to Rocket Lab’s Production Complex for refurbishment

Has anyone been able to economically refurbish engines fished out of salt water? I know that NASA and SpaceX both failed at this, why would the Electron be better suited for saltwater recovery?


Maybe because they don't have the most vulnerable part of a rocket engine -- turbine?


1. Is this possible for Election and not for, say, Falcon 9 due to the difference in mass?

2. Is that a retro burn in the animation or reentry heating?

3. Are they gonna catch those discarded battery packs too? Should be easier than a whole first stage.

4. If they nail this, catching everything by helicopter shouldn't be that much harder.


1. yes - they don't have the mass for the fuel to get back 2. reentry heating 3. probably not (I think it's the 2nd stage that drops its batteries) 4. because they don't have the mass to do the return they don't have a lot of options


I suspect they will modify the design to not drop battery packs anymore.


I suspect the battery packs cost basically nothing in comparison to the rest of the stage. (I'm also not sure that the batteries are even reusable - if there's ever a time to sacrifice rechargeability for performance, it's here)


(and I'm pretty sure it's the upper stage batteries that they drop)


The helicopter in the animation looks like a EC225 or AS332 "Super Puma"


He wants to do it with Super Puma. Ahem, its maximum takeoff weight is 9 tons...

There is a limited supply of Mi-26s in the world, and I bet nobody these days will be eager to even lease theirs.

Why none out of those "new age" launch companies uses the simplest solution? Parachutes and solid fuel rockets for soft landing.


How does one throttle a solid fuel rocket? And what does the thrust profile look like in the seconds leading up to, and after, ground contact?

If you are really interested, try doing suicide burns in Kerbal Space Program. KSP is great for disspelling "why don't they just" questions.


1. Calculate terminal velocity of an empty rocket on a parachute

2. Calculate thrust and fuel reserve of solid rockets to reduce the speed of the rocket stage just below the speed at which it can survive hitting the ground

Such schemes were used in Soyuz for half a century, Zenith for 30 years, and other parachute landing system for tanks and other military hardware.


a) because they launch out over the ocean

b) because you can't turn a solid rocket off once you've landed

c) and because of the rocket equation any added mass costs so much more, if they could afford the mass for solids for landing they could just add more liquid fuel and do the spacex thing, that would likely use less mass


a) rocket should have huge positive buoyancy

b) you just have to size them right for the rocket mass and terminal velocity with a parachute

c) definitely, but propulsive landing also does not come with fuel as the only mass penalty. You also have to add mass of heavy hydraulics and sensors into consideration.


The SpaceX reason for doing barge landings rather than just fishing the rockets out of the sea is that salt water is very damaging to the rocket. Landing on the barge requires active directional control.

Most of the Soyuz deceleration is done with parachutes, I'm not sure how that scales to a whole first stage.


yes, exactly, salt water is VERY VERY bad

Your main engines are gimballed for guidance, that mass cost is already paid


I hardly can imagine why.

Their rockets are carbon fibre, aluminum-lithium, and steel alloys with such overkill corrosion rating that they should stand strong acids, not simple seawater.

And mass cost of hydraulics for propulsive landing != hydraulics for regular gimbal steering. Those should be very different in their power and mass. Not only they have to steer the gimbal much faster and harder, but also aerodynamic elements on the rocket, fuel throttle, and those legs they land on.


The expensive bit is all the electronics and turbomachinery in the engines.


So to clarify, you want them to haul up a solid booster that has enough thrust to slow the rocket down during reentry and then deploy parachutes?

Just the fuel for retropropulsive landing is already a major blow to falcon heavy or 9's payload. How much of a payload hit do you think this will be?


No, no, no, the order is reversed: balutte/supersonic chute > regular chutes > last second breaking impulse with solid fuel engines to slow down the stage just enough to withstand the landing.

Just like the landing systems for airdropped military hardware.


Is this really the most simple available method?


Basically, yeah. And it is pretty simple and it has been done before by Exos Aerospace (formerly Armadillo Aerospace of John Carmack): https://www.youtube.com/watch?v=_G9dJ3IQQVk

... the helicopter recovery of stuff falling by parachute from space bit has also been done before with the Corona capsules: https://www.youtube.com/watch?v=Q2YQqAnEN_0


I spent a solid minute thinking this headline was about Electron the app shell and Rocket the rustlang web framework...


Yeah, I was honestly surprised this was about a real rocket, and not some sort of effort to get electron more efficient.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: