Hacker News Reader

Comment by btilly | Hacker News Reader

td;lr

Existing SpaceX rockets land on legs.

For the SuperHeavy booster that will launch Starship, Elon wants to take the legs away and instead have it land on a platform intended to catch the rocket on its grid fins near the top.

This requires a very precise landing. But gets rid of a very heavy piece, and makes it easy to build a shock absorber. The downside is that you'll need to replace platforms any time one of these misses, hence the need for a very precise landing.

(The landing already has almost no margin of error given that they land with a suicide burn.)

View on HN

Starship and SH are (AFAIK) meant to be able to hover (more thrust/more throttle-able than F9) so the suicide burn shouldn’t be an issue

this seems to be what could make this a possibility.

it'll take some work but the video highlights this strongly: falcon is light-weight & with modest throttling. it's gonna land, and just barely drop velocity, wherever it was pointed. there's so much thrust to weight & not enough fine throttling that running the engines for a bit isn't gonna help let it position itself.

ss & sh are heavy as heck. as much as throttle-ability my gut feel is that it's really the mass that helps make hoveribility-positioning sensible. some ability to get to the area & correct, & then land. there's enough mass that a bunch of burning isn't going to immediately blast the craft away again.

that's what I got from the video, at least.

I can't find anything definitive on that.

However looking at Wikipedia, it seems that Starship is supposed to weigh around 500,000 pounds empty, and will have 6 raptors. Each raptor can produce between 200,000 to 500,000 pounds of force. So it is theoretically possible to hover on them.

However Elon wants to land on Mars. There a single raptor produces more thrust than is needed to hold up the ship. So you're back in suicide burn territory.

But Starship is supposed to have 6 raptors while SH to have (in current designs) 37. So there is hope that

Starships probably won’t be landing on Mars empty. They will have up to 100t (220klb) of cargo, and maybe some extra fuel.

Even at 0.4G that should be more than 200klb, they’ll probably need two engines to hover.

If they're landing on the launch platform, they've already solved the backpressure issue from the rocket exhaust being that close to the rocket body, haven't they? I'm not sure I see how the catcher's mitt helps. Didn't the legs and platform already have to hold the static weight of a fully loaded Heavy plus a fully loaded Starship, and now it just has to hold up a mostly empty Heavy plus whatever delta V or delta H at rocket cutoff... Is that more than the empty/full weight differential?

I think they're not landing on the launch platform. With catcher's mitt on a different platform, launch configuration won't have rocket legs, just a platform tower holding up the rocket.

what makes you think the "catchers mitt" can't/won't be another couple meters up?

The problem is that when the rocket shuts off, they usually haven't quite landed right. If they are still moving, the legs have to catch the rocket. If they are still a bit in the air, you'll fall down and have to catch the rocket.

The legs on a Falcon have built in shock absorbers. Here is why. The energy with which you hit the ground is 0.5 m v^2. This has to match force times distance while the legs catch it. If it is perfectly rigid, the force is insane, even when rigid. But if the legs can flex, you can make this work out.

But if we've got a structure that catches you, then the rocket doesn't need to have those shock absorbers. The structure does. And if the longer the distance over which the shock is absorbed, the smoother the landing.

Now in Elon's plans, Starship needs to be able to land on Mars unassisted. So it actually needs legs with shock absorbers. But SuperHeavy only needs to take off from and land on Earth at pre-arranged locations. So he can make it more efficient by not building in any unnecessary shock absorbers.

Side note. This is why seatbelts matter. With a seatbelt your crash is spread over all of the time that the car is crashing. Without it your crash is spread over the time it takes your head to hit the car. When your head stops in 1/10 the distance you have to have 10x the force. It is not hard for this to be the difference between being shaken up and having your brains scrambled.

Super heavy can hover on reentry, so it will land far softer than a Falcon 9 first stage that is require to do a suicide burn.

This is also true of Starship on Mars since it will be massing over 200 tons.

1. The catcher’s mitt can suspend the rocket far higher above the ground than the landing legs can, reducing backpressure. Even if this is already “solved”, this could make that solution cheaper and easier.

2. Right now, the legs need to be capable of absorbing the shock of the landing. With a dedicated platform, not only can the weight of the shock absorbers be added to the platform rather than the rocket but you can also potentially absorb over a dramatically greater distance and time (increasing the proportion of successful recoveries).

Literal impact (the engineering term of art) is usually the controlling force for these types of structural calculations. The force varies with respect to velocity-squared, and due to the dynamic nature of landing, the forces are not likely to be distributed evenly (high pressures at unpredictable locations), so you end up with the controlling condition needing to be much stronger than perhaps even supporting the (static, evenly distributed) weight of a fully loaded vehicle.

Sounds a little bit like the soviet/russian(and maybe chinese derivative?) lock mechanism of their launch towers, but in reverse. Why not? Aren't the grid fins of russian origin, too? :-)

Is suicide burn a viable landing method for manned rockets? Would need a comfortable G force, perhaps 3g for 10 secs?

And then how many successful unmanned demonstrations until passengers feel comfortable trusting it? 1000?

I think that unless they prove they can nail this every time, it'll be tricky to use a suicide burn for landing manned craft. Passengers may also object to the name of the maneuver.

I guess a manned starship is many landings after the first unmanned landing, at least in Earth.

> manned starship is many landings after the first unmanned landing

What number makes you comfortable? 100 flawless landings, 1000?

My guess is, after about 20 successes in a row you start to have competent people that want to risk it and go to Mars. Being first at that carries huge prestige.

If your minimum thrust is > 1g then you can't hover and so have to use a suicide burn. Manned rocket launches have a sustained 3g force. Rollercoasters commonly hit 5g.

There is some room in suicide burns between what humans find easy to handle and what is possible to make the rocket work. But you do have to think about this.

Viability of ~= 1g thrust? Perhaps keeping the same engine some downwards thrust can be redirect sidewards with some obstructing shape moved into the exhaust plume

Easy. Don’t miss.

Reply on news.ycombinator.com