Can SpaceX Catch a Starship Super Heavy? (Like Falcon 9?)
ฝัง
- เผยแพร่เมื่อ 10 ต.ค. 2024
- SpaceX has mastered Falcon 9 booster landings, but how accurate are they really? Join NSF's John Galloway as we dive deep into over 100 landing videos to analyze just how close these landings are to a perfect bullseye. Plus, can this data tell us how accurate a Super Heavy booster catch attempt might be? Find out as we explore SpaceX’s potential for catching the massive booster using the Starship program’s chopsticks. Don't miss out on the mind-blowing details!
🔗 NSF Store: www.nasaspacef...
⚡ Become a member of NASASpaceflight's channel for exclusive discord access, fast turnaround clips, and other exclusive benefits. Your support helps us continue our 24/7 coverage. Click JOIN above to get started.⚡
🤵 Hosted by John "Das" Galloway (@KSpaceAcademy)
🖊️ Written by John "Das" Galloway, Alejandro Alcantarilla Romera, Thomas Hayden.
🎥 Video from D. Wise, John Galloway, Jack Beyer, Max Evans, SpaceX, ULA, NASA.
✂️ Edited by Thomas Hayden.
💼 Produced by Kevin Michael Reed.
🔍 If you are interested in using footage from this video, please review our content use policy: www.nasaspacef...
L2 Boca Chica (more clips and photos) from BC's very early days to today.
🔗 forum.nasaspac...
(Join L2 and support NSF here: www.nasaspacef...)
#SpaceX #Falcon9 #RocketLanding #SuperHeavy #BoosterCatch #Starship #SpaceTechnology #RocketScience #SpaceExploration #SpaceInnovation #NASA #NSF #Engineering #SpaceXLaunch #BoosterReusability #RocketAnalysis #ChopsticksCatch #SpaceTech #FutureOfSpace #ElonMusk #Aerospace #Falcon9Landing #LandingAccuracy #RocketLandingAnalysis #SpaceXFuture #SpaceTechAnalysis #SuperHeavyCatch #StarshipProgram #SpaceXBooster #BoosterLanding #Spaceflight
Das is a national treasure.
…eh… International actually‼️
Here here 🍻
I’m German and I read that in Arnold Schwarzenegger’s voice. Very funny
@@oljobo Even better ...!!
I am the guy's die hard fan ..!
The other fav of mine is Ryan Caton.
Both these guys rock.
One key difference to remember between Falcon 9 and Superheavy (aside from the droneships being on a heaving sea) is that the Falcon suicide burns into the landing with very little room for error. There is a point in its deceleration profile where the booster absolutely needs to be in contact with the deck and the engines off. The Superheavy (theoretically) can slow down, hover and adjust its position, much like New Sheppard and (theoretically) New Glenn, so, in theory, the margin for error can be larger, at least as much as the landing propellant reserves allow.
Yeah exactly. I'll have a couple seconds to correct it's position, whereas the F9 basically has to worry primarily about getting it's speed to zero at exactly the same time it reaches the drone ship, or it flies back up. With starship they have time to make corrections (however it'll be roasting the pad during that time so probably don't want to spend too much time doing it (but it'll just be one engine at half throttle not 33 at full throttle as well)
@@takanara7 Indeed, someone else noted a three engine landing but that's not required, the landing is possible with only one engine and it was raised to start with three engines and then shutdown the extras once the engine expected to do the landing is shown to be working properly, otherwise one of the other engines will be picked.
Imagining that big booster falling from the upper atmosphere and then hovering and maneuvering into a catch on chopsticks is mind blowing. So excited and glad it's happening Sunday. This is just an iteration, tho.
Yeah I'd assumed this was the caveat being held to mention at the end.
Falcon 9 is essentially landing like a glider (suicide burn equivalent to air brake) compared to the powered aircraft decent superheavy should do (though it's not got 'go around 'capability!). Offers the adjustment capability F9 just doesn't have.
I do worry they choose an engine to use and it cuts out while adjusting though 😵
@@craigwelsh I'd imagine that's why they'll light 3 at the start of the landing (is it a landing if it's getting caught?) and then shut down based on telemetry to give it the best performing engine for final catching/landing control
Ok here is a list of all the problems I have with this video.
1. It ends after about 17 minutes.
Thanks NSF for making this awesome content! ❤
I've given and attended many briefings and seminars in my live and no one does it better than Das. He's knowledgeable, prepared and personable all at the same time. He could brief a cooking show and it would be just as interesting. Thank you Das!
Great Analysis DAS! One thing..if you notice,the last 100 or so Falcon 9 successful drone ship landings the flacon 9 always lands in the same compass orientation so Spacex has been practicing for this tower catch orientation attempt with Falcon 9 as a surrogate for a long time. Given no unexpected hardware failures with the booster and chopsticks I think SpaceX has a good chance of success.😊
Considering the accuracy of last flight's booster, my concern at this point are the chopsticks, they're so wobbly lmao
Best of luck B12!
Also, thank you for the metal pipe sound effect. Made me laugh way too much.
I saw that someone claimed 0.5cm of accuracy and I’m wondering if that’s a typo. Since it landed in the ocean and not on easily measurable land, they must be using GPS to measure that and to my knowledge the current GPS system itself isn’t that accurate so it doesn’t really make sense. I’m wondering if they meant 0.5 meters. But even it is was meters instead of centimeters that should still be accurate enough for the chopsticks. Very interested to see how the actual landing plays out, unless they abort to the gulf of course but we’re still going to see a catch attempt sometime soon…
I read that public GPS isn't that accurate but other forms coupled with different positioning systems can be that accurate. @@FerociousPancake888
@@FerociousPancake888"someone" that is Bill Gerstenmaier
Why do people have this weird notion that it's magically less accurate to measure out in the ocean because it's not land (or that GPS is somehow an issue??)
The booster knows where it needs to land, and it knows where it is, it's not hard to find out how off you are from there. Even then, the landing site probably had multiple buoys and they could've mathed it out using positional data from them
Elon said the inertia in the chopsticks is a problem, that's why the next tower is going to have shorter arms.
same man
Das seems to imply that landing the Super Heavy accuratly will be more difficult than Falcon 9, largly because it is heaver. But the Super Heavy has a lower thrust to weight ratio, a effectively stationary target, and better controll due to three engines with TVC instead of just one. The Super Heavy should be signigantly more accurate than the Falcon 9.
😮here’s hoping you are right 😊
Doubt it.
@@kokomo9764wut?! It just facts, you can't just say "uuhm nah I don't think so 🤓☝🏻"
Plus Raptor can deep throttle much further then a Merlin. SH could even come down beside a tower and hover over to the arms if there's enough fuel - we saw it hover above the waters surface already.
I like that, but one thing I think could be an issue, there is no re-entry burn, so it will start landing burn at a high speed which I think could throw things off. Not saying they won't do it, but dang.
I think the reason SpaceX doesn't make Falcon 9 land more accurately isn't that they can't, it's that they don't need to, because anything that lands on the drone ship is basically equally successful.
bingo
Mire than that if F9 boosters landed dead in the center every time that spot on the droneship would be destroyed after a few landings
@@ДинозаврЧелик I'm sure they would repair / replace the landing surface long before any real damage occurred to the drone ship
@@Pegaroo_ If you land in a random spot of the landing drone you will have to do less repairs overall.
@@ДинозаврЧелик Sure but even if it did land in the same exact spot every time they are going to repair the deck rather than let the drone ship be destroyed
The biggest difference is that the falcon 9 cannot hover while the booster can. This nullifies this thought process because the adjustment capability is drastically larger
Starhopper for the win lol
It doesn’t nullify anything. If you believe super heavy has better control authority, this analysing Falcon 9 give a good idea on a worst case accuracy distribution. From this, it should be very clear the likelihood of an accurate landing is quite high. Actually catching it is a different story lol
@lanefaison6257, I am no engineer.
But if the booster pins are fixed on a rotatable ring, and the rotation is enabled(electromagnetically?)by the Chopstick, would it make the pins landing exactly on the catch rails?
We might need a max of 20-40 deg rotation, basically a twist to land the pins on the catch rail.
@@guppi277 nice idea in theory, however that’s a whole lot of additional weight and complexity to the booster, not to mention the fact that it needs to be strong enough to hold the weight of the booster, which is unlikely.
If SpaceX was concerned about not being able to control booster rotation well enough, they would have simply added extra pins instead.
@@guppi277would be easier to design a handful of more fixed pins at the circumference of the booster. Lighter, no moving parts, no sensors required. Just the mass and drag of some more pins.
The fact that they are listing it as the primary objective makes me confident that they will get it. So far they hit every primary objective during the previous flights...but that doesn't make it any less crazy.
If we can do these much analysis then just think how much SPACEX engineers can!❤ Nice Video
Will be an amazing, butterflies in stomach moment!
Sub-centimeter precision is better than I can do with a Sharpie.
the problem is not landing but getting a crane big enough to put it on a transporter big enough to get it back to the launch site. the chopsticks might seem like a crazy idea, but landing a super heavy on a boat and then picking it up to put on a transporter to trundle back to boca chica is kinda INSANE.
i was feeling pretty confident about the catch until now, my update (!) is that either it will go perfectly this time, or they rip the chopsticks off. i can't wait to see it launch, coming back will be the icing on the cake. and for goodness sake elon, stay out of politics.
@HarryNicNicholas Chopsticks can't be ripped off. We've seen them handle 700 tons. Even with the force of Superheavy coming down, it will be in worst case the Booster that gets ripped, not the chopsticks.
The OLM and Tower too, are way sturdier than Superheavy.
They should build a giant drone capable of lifting the rocket. Impractical, sure, but also extremely awesome.
@@takanara7 ??? Impractical? That technology doesn't even exist currently.
@@realEpicGold is that so? That's smart design, then. The most expendable components should of course be the weakest links. I heard that some of the reputation of German WWII tanks for having bad transmissions was due to, I believe the final drives?, being designed as the weakest link to prevent more expensive components from breaking. Such a method does, however, mean that improving the resilience of the system requires improving the resilience of every single component.
So if Super Heavy turned out to need reinforcement to survive rough chopstick catches, that would likely also require chopstick and maybe even tower durability upgrades.
I am torn on this one. It's ambitious, but we've seen SpaceX deliver on ambitious before, so we know it's possible. I think if the booster can return to the tower and guide itself in to the chopsticks at all, then there is a chance that this 'just works' thanks the the extreme dedication and work of the SpaceX and Starbase workforce. For once, this literally IS rocket science, and I am hopeful that it just works as planned. - If not, I do worry about the consequence of not completing this mission successfully and what that could look like. But one thing is for certain, if this ends in a big fiery ball, its going to be spectacular and SpaceX is simply going to rebuild it and try again. Best of luck to all teams working on this epic project!
I was very 50/50 on a successful catch up until it was stated that Super Heavy landed within half a centimeter of accuracy. That alone pushed me to 80/20 success
Making the impossible merely LATE
Catching it with the chopsticks just sounds absolute bananas to me. If something goes wrong very late the OLM could be damaged. But then again, when SpaceX says they'll land their F9 booster and fly them again, everyone thought they were crazy too. SpaceX will make this work I'm sure, but this flight 5 catch attempt is going to be very hair raising but I'm hoping for the best.
10 years ago landing rocket boosters was bananas, now it happens every couple of days and is only newsworthy when it doesn't go perfectly
- Why they use these pins and not a ring is a riddle to me. With pins the odds are very tiny.
- Probably they have a FSD for the chopsticks. Install a camera that observes the booster position and adjust the stick speeds accordingly to hit at the same time. Including some variations in the position while he is hovering and waiting for the closing.
I was hoping someone else would bring up the rotation necessary to align the pins. Why is this not one of the biggest concerns?
@@craigbarboza7992 Superheavy has a TON of roll control. Having 3 gimballing engines makes it very easy to spin.
@chandlerday6471 agreed, but it also has multiple engines for putting it in the target location. It just seemed like an oversight to mention misalignment of the pins.
Answer, yes. Thoughts personally: I think there’s a 50/50 shot the director says go for catch. If so, 40% chance of full failure, 30% chance of it catching on the grid fins, 10% right on the pins. Those are my bets and I’m sticking to it.
And the remaining 20%? Lol
It won't be caught on the grid fins they are designed to help steer the booster but they don't have the strength hold the weight of the booster.
As for the other 20% successful catch but the booster had to hover in place too long and causes damage to the tower and pad that will delay the next launch by months
The Super Heavy boosters are larger and weigh much more than a Flacon 9 booster. That gives them more inertia which means they are a lot more stable in the air when landing.
I think the landing will be within a funnel that gets smaller the closer the booster is to the chopsticks. If the booster is not inside that imaginary funnel by the time the 3 landing Raptors take over the booster is aimed at sea.
Also, the 3 Raptors are throttled down so should one of the 3 fail the 2 remaining can throttle up (and gimbal) enough to compensate.
I really think they will nail it.
An engine failure during landing could spell doom for the booster. Landing even slightly lopsided on the chopsticks could mean a much more intense impact.
If they leave enough fuel to hover it in to place, that's also enough fuel to fire enough engines for a very short parabolic and ballistic flight from tower to sea.
@@PiDsPagePrototypesthat would be an amazing sight if it did that. No idea if it can though.
@@PiDsPagePrototypes No, that would need a reverse acceleration and would take much more fuel than they could have left, IMHO.
So, if you have more inertia you can maneuver more precise? I would say the opposite, this is more like parking a huge truck in a tight parking lot, a smaller car can make more precise parking and if something get wrong less inertia means that you can correct faster, also for a precise insertion probability they need some powerful thrusters on the upper part (which they don't have), but after few failures maybe would put some.
It is actually a good analysis whit out having hard data. Good one.
Dude, I think you did a nice job explaining this.
I've heard that SpaceX intentionally scatter Falcon 9 landings around the deck of the droneship to avoid excessive wear in one place from the engine exhaust.
No idea if that's true, this is just hearsay, but it does sound plausible if their accuracy is anywhere near as good as Bill claims.
Given the distribution looks pretty normal I'm not totally sold on the idea (though I've also heard it), but it's definitely not something we can rule out!
Not surprised to see so many comments. Yes Das I learnt a lot. A focused; detailed; researched and easy to follow presentation I appreciate the work that went into it.
The fact that the two rockets are so different in size makes things easier. Compare the ride quality of a 747 VS a Cessna 162 or a zodiac vs Cruise ship
Which one is easier to park... a zodiac or a cruise ship? 😅 Das
@@NASASpaceflightthat depends on the captain 🤣😉
Can't wait to see this attempt. We're headed to SPI tomorrow and setting up remotes Saturday. See ya there, Das!
Thanks for the best technical overview I have seen in years on TH-cam.
I think they are going to be about 75% successful. At worst maybe they will just kind of crush or dimple the sides of the booster but I think one way or another the chopsticks are gonna be able to grab It.
But what an amazing amalgamation of parameters that have to be dealt with.
The animations make it look so simple.
My biggest question is why are the pins so small, Like why aren't they more fin-like, This would helpThe chopsticks have more surface area to engage in as they slap the booster.
I am a heavy construction civil engineer with 50 years of experience building every type of structure that has been built-in the US, and you made the presentation so easy and personable to follow along and it's great to acknowledge assumptions. The picture graphs really helped fortify your precepts,
Thanks again for the best I've ever seen on this subject.-Ken-
Awesome analysis Das and NSF - Scott Manley would be proud
Great video - I've been thinking about the accuracy needed for a landing of Superheavy for a while. I can't understand why SpaceX have made the pins so small or at least have some system that goes all the way round the booster to simplify a catch. I actually think Gerstenmaier might have made a mistake/misspoke when he said the last booster landed with 0.5cm accuracy - it seems much more likely to be within 0.5m accuracy which aligns to what we see with Falcon 9 landings. Not sure how you could even measure to 0.5cm accuracy out in the middle of the ocean without a solid/firm fixed reference point
John, one thing that Booster has going for it is the ability to hover, whereas F9's landings are all hoverslams. This should give the system time to minimize any positional errors, and possibly roll errors using the 3 engine's TVC.
About that hover... while it's hovering it is blasting the top side of the OLM where water deluge can't reach. I don't think they will want it to hover very long to keep the damage to the OLM light-ish
It hovers with only 3 engines I think. So the flame will not be comparable to the launch. Still a problem.@@minibeefcake
@@minibeefcake Three things come to mind: 1.) 3 center engines, not 33. 2.) Those 3 engines won't be at full power. 3.) The booster will be elevated quite a distance from the OLM (if it's even over the OLM for landing, IDK). Energy density dissipates with the square of the distance (I think), and thermal energy similarly.
Hope these dance partners meet up smoothly when the real flight landing tests begin. I don't want the dance music to end in a record needle scratching sound. I am glad you showed off all the complications of why this next landing system will be so challenging. Enough of people saying "why don't they just do this or that?" They have explored every option and they have the best minds working on the challenges. They will flesh it out
I think an interesting thing to remember or think about is that, SpaceX can place extremely accurately located transmitters around the tower so that the booster will know its exact location Way more accurately than just GPS alone. Once the booster is near the tower, it very well could know its location within half a centimeter.
The one big advantage that SpaceX has with the Superheavy is that it can hover, the Falcon 9 cannot, so the Superheavy can have less velocity as it is approaching the tower and stay at a given altitude while it's thrusters rotate it.
can you do some analysis of the flight 4 booster landing video and falcon 9 landing video and see the timing/velocity of them?
Great Analysis that Made the problems of catching visual and understandable. Thank you!
The biggest difference is that the Falcon 9 is doing a suicide burn, whereas the Superheavy booster can do a controlled hover. The F9 has to hit its target like a dart hitting a dartboard, getting it perfect in one shot. The Superheavy can work with the chopsticks in concert to shift horizontally and even adjust for some degree of pitch/yaw when the chopsticks make contact. As you noted, the roll axis is an issue because they only have a few degrees of roll where the pins hit the catch pads. They also have the least control capability in the roll axis due to the nature of using gimballing engines for control, so that's likely to be the biggest challenge.
if you can thread a cruse missile through a 3 foot window from a thousand mile away ...!? i don't think it would a problem ,
@@oldboy707 Exactly. Other than roll, this should be easier than the F9 booster landings, and they have those under control.
I thought the drone ships have +/- 2m position accuracy in the ocean (presumably correlated with sea state). So F9 accuracy could be notably higher wrt to precise coordinates they are trying to hit, as opposed to where the drone ship actually is. I have no idea though whether SuperHeavy has better/worse/similar accuracy to F9.
Yes! The droneship isn’t perfect and Falcon 9 landing doesn’t need to be as precise. Why optimize landing accuracy at the expense of other landing parameters.
Additionally to get centimeter level landing accuracy you will likely need some sort of navigation enhancement. Maybe differential GPS or some other augmentation from the ground.
It's quite a dynamic system, having to correct for ocean current, ship velocity etc. I'm sure they try to keep the drone ships as neutral as possible relative to the earth surface, but between swell/waves/currents there will be some degree of dynamics at play that will not needed to be accounted for with a chopstick landing. That plus the high control with hovering, it seems likely that can probably do better than Falcon 9 drone ship landings assuming the flight trajectory is "norminal".
Always said this. Droneship is attempting to match a GPS location. As is the booster. I think you'd have to see how off the droneship was at any second. RTLS is like Starship, a fixed target.
My belief is that landing accuracy is a function of not only the technological limits, but also the amount of spare fuel available for the landing burn, and the ability to do a hover. F9 is being pushed to the limit, leaving little spare fuel for correcting minor lateral deviations during the landing. F9 also lacks the ability to do a hover, as the central engine on the booster cannot throttle down far enough to hover when the booster is on low fuel. Thus, F9 has to perform a hoverslam.
Basically, I'm saying F9 boosters land inaccurately partly because they can, and partly because of the engine layout. When considering Super Heavy, it has the ability to hover, allowing it to make a more cautious, slower approach to the landing site, given that it also has more spare fuel, which should also be possible given the Starship is running no payload. Therefore, Super Heavy should be able to achieve much higher landing accuracy than F9 normally does.
My main concern is those chopsticks. On one hand, it's concerning how slowly they close, on the other, after seeing how much the tower wobbles, how bursty the movement of the chopsticks is, and how they plan to just slap the booster, I'm kind of glad the chopsticks move so slowly. Provided that the booster has enough fuel to hover over the pad for 20 seconds, I think this is very doable.
84 is quite a good sample size. Pretty good analysis based on your data. Surely SpaceX has way more info and control.
An amazing Das explainer video. Thanks again Das amazing. Great work. And thanks NSF team for all the hard work in the background for this video. Amazing analysis. Keep up the great work.
Well its gonna be a sight to see which ever way it goes.
It seems to me that as soon as the chopsticks impart a roll, the engines will gimbal to counteract that force. When they do it will push the rocket in the direction it needs to go. The rocket will remain vertical and move sideways. Under engine power.
You have to also take into consideration water movement on the done which makes the falcon9 drone landings even more amazing
yea as others have said, the "hover slam" is a very different landing style than Boosters ability to actually hover. I think its a matter of how much of three(ish) raptors firing next to it the tower can handle, while the bootser/chopsticks adjust. but i think the tower can withstand a minute or two of raptors hovering next to it if it really needed, but likely a few seconds for things to adjust is all it would need.
With respect to the likelihood that the booster being off centre and therefore one chopstick would engage before the second chopstick engages, therefore requiring the first chopstick to move the booster towards the second chopstick, you have to consider that the rocket engines (however many of the engines that would be required to be operating ), to maintain the booster’s height and orientation while performing the “catch”.
Excitement Guaranteed
Thanks for some great analysis, guys!
The thing is, Falcon 9 _HAS TO_ land at a more or less predetermined time during the landing burn, the moment when its velocity hits zero. It doesn't have a choice, because its one-engine TWR at _minimum_ throttle is greater than 1, meaning if it doesn't shut the engine off it will start _going back up._ In other words, Falcon 9 is _incapable of hovering._ It doesn't have the time to make its landing even more precise on the horizontal plane -- as long as it's _close enough_ to the center of the landing site when velocity reaches zero, that's all it cares about.
Superheavy _will_ be able to hover. It will be able to take the time to come in a little more carefully and hold a constant vertical speed (that is, a slow final descent) while adjusting for any inaccuracies in horizontal position, not to mention (as he showed in the video) the "chopsticks" can swing side to side and there's wiggle room along their length as well.
Two comments: 1) there's a lot of movement on the ocean so the error in landing on the ships appears to be much worse (at least 1m I'd say, depending on the wave state). and 2) Falcon 9 does a "hover slam" (a single merlin is way too powerful for a nearly empty first stage) but the Starship booster has more control because the thrust to weight of a single raptor to booster is more favourable.
Really interesting and informative video. I hadn’t even thought about the landing pins and the constrains on roll this would cause.
Apologies if I missed this, but something to factor in, Falcon 9 drone-ship landings have to contend with sea conditions during the landing. Swells could cause the ship to move a meter in 3 axes, and that’s not even considering the landing pad might not be perfectly horizontal. This might mean the accuracy of the F9 is actually better than that calculated by analysis of landing location on the ship.
The starship booster can hover and the chopsticks can theoretically adjust position both left-right and up-down. If you get down the software for this it could potentially become even more reliable than landing on a stationary pad ever could. In a way, you get some additional redundancy since there are now 2 things that can adjust to the landing instead of one.
Fantastic video! Hopefully the tower survives the first catch.
Great video Das. This answered many questions that I have had running through my head.
Wow. I knew this was gonna be complicated but didn’t realize how many variables they’d have to account for, including roll. Great analysis. Can’t wait to see how close they get on the first try!!!
That was awesome dude! Keep pumping them out. And don't change I think saying 💥🚀
🤘😜🤘
I learned a few things from this video, including the fact that the bumpers are hydraulic... which I didn't know. Thanks for sharing!
Definitely the best thought out guesstimation I've heard. Anything is possible, excitement guaranteed 😉
Excellent analysis. I especially love that you were able to get a normal curve. Good job. Very informative
Appreciate your analysis, I think you did the best possible with the data you have. It was definitely entertaining and interesting. We will have to see what happens in real life.
Wow Das how did you fit working all that out & being out in the field setting up camera's. Yes I learnt a lot again ....Don't forget to order your flight 5 patch folks...Sadly Das not included 🤣🤣 If you know you know
the only exception to the analysis that I can think of is the drone ships are moving in 3 axis while on the ocean most likely changing the landing location on the ship slightly for each flight. I'd have liked to see an analysis of the RTLS boosters since that is a fixed location to land, not as much data to analyze but is a better like for like comparison to what is being attempted.
Got a good laugh out of the booster clang😂great video! Let's fly this beast and do the unthinkable!! I need more SN8 and Flight One-Style excitement🎉
Its a whole other beast. I do think on shore landings are the better thing to study since the drone ship itself is also moving with ocean force. But with the booster able to hover, it has way more room to correct than a suicide burn
Okey, watched the entire video. Now I can comment.
A few points to consider:
- First, we know the drone ship isn’t a stationary target. The original drone ship was said by SpaceX to be capable of about 3 meters of stationkeeping accuracy. The newer drone ships are likely better, but it’ll always be a moving target.
-The assumption here is that SpaceX is always aiming for the dead center of the pad. Even if the drone ship wasn’t moving, it would be a good idea to aim each landing to be slightly offset from the previous ones to minimize wear on the landing surface.
-Physics is on Boosters side. A taller, heavier vehicle is by definition more forgiving to control. Not only is it less affected by external influences like wind, but an object with a lower fundamental frequency of occultation is much easier to control than one with a higher one. Like balancing a pencil on the tip of your finger vs a broomstick.
-Lastly, the booster can hover, whereas F9 can’t and needs to perform a (still incredibly impressive) hoverslam. Even if the booster is slightly off, it can just hover until it gets where it wants to be. The way we currently imagine the landing will go doesn’t quite include a hover, but it does include a final decent speed much lower than what F9 can’t achieve.
On a sidenote for those skeptical about whether or not sub-centimeter accuracy is even possible, it is. RTK-GPS using a base-station is already centimeter accurate without using the military GPS frequencies (which SpaceX may or may not be using). Combine that extremely accurate GPS with the also relatively accurate IMU, and all the advantages Booster has over F9 that I mentioned above, and sub-centimeter precocious really isn’t all that unbelievable.
There is a huge difference between a Falcon 9 return to launch site and a Starship Super Heavy Booster a return to launch point. Falcon 9 launches from one point and lands at another point, basically following preprogrammed point to point coordinates. Whereas Starship Super Heavy Booster will launch from one point then come back and land at the exact same point. Now, obviously the Earth never stops spinning and technically that spot will have moved but the vehicle is moving in relation to a single fixed point on the globe which is far easier.
Hi Das. One variable not mentioned in the F9 landing accuracy is where Spacex measure the booster's position from. If it's from where the gps box is, then it's probably near the top. Depending on the tilt of the booster and the roll of the droneship, two identical positions of the booster would land on different parts of the deck.
I'm an OLD engineer.... this stuff left me behind 30 years ago. BUT, I'm still interested in it... Today, I'm just another nerd with a worthless degree....
What happens if the grid fins are caught? Will they hold?
The flaw I see this analysis is that it assumes that the drone ship is perfectly station keeping. But I've heard (I think from Berger's new book) the ship can drift +/- 3ft from its GPS hold point. Your analysis assumes all of the landing error is due to the rocket's trajectory, so it over estimates the trajectory error. Since Stage 0 is fixed to the ground, the rocket trajectory will yield less error.
I have to assume the guy meant half a meter not half a centimeter. I don't even know how you could measure half a centimeter accuracy in the middle of the ocean.
Military GPS is far more accurate than commercial GPS and I would expect that SpaceX gets access to the military grade instead of having to use the commercial grade.
Yeah, 0.5 cm sounds like an impossibly precise landing. (I also doubt military GPS has 0.5 cm accuracy.)
When map predicting we used to use 6 figure grid accuracy for mortars which is 100x100m “accuracy” which was good enough but in reality once we had lasers and MGRS (GPS) we could go to 10 or even 12 fig Gr which was ridiculous as human errors in firing produced inaccuracy making it pointless. Ten fig Gr is the best I’d bother with for say hitting a trench.
However for an experiment we used 12 fig gr once to hit a dead tree branch. It hit smack on. The smallest we use with mapping is 14 figure grid but nobody ever uses the last two digits as it’s pointless for human map reading.
Military GPS is encrypted and more accurate but can be degraded by various reasons. I too have no idea how he calculated 0.5cm accuracy. Even 0.5m is fantastic.
Great analysis, great work well done! Thank you!
Before I watch this video beyond the 20 second mark I have to note that Falcon 9 lands extremely accurately for a system where the landing is only "powered" to slow down the booster so it can land, the power of the engine is used only to stop the booster, not to guide it. The location the booster lands is controlled by the grid fins. This is a consequence of the inability to throttle the Merlin engine down low enough to allow a powered landing. Starship has the Raptor which was designed to be highly throttleable and allow steering using the engines which is a reason why it took some tries for SpaceX to master that form of landing.
Looking forward to seeing if you note these differences or not...
And no, the difference between a grid-fin controlled accuracy and a grid-fin plus steerable rocket that can also hover was not brought up. You did note how very close the grid-fins alone could get a rocket though. I still like the video overall. Just a few days now and we'll all know how that's going to work out on the first try, so great timing on being ready to discuss this.
Well i hope Space X has reached out to you guys for your helpful input of this video ❤️ And i agree DAS your a legend! Keep up the good work NSF team 😊
Regarding roll control, two points:
1. Navigating to the landing requires complete control of roll to make any trajectory adjustments.
2. The drone ship generally appears at the same angle to the booster-mounted camera and that's a possible way to assure yourselves from those 84 assessed landings that they are indeed in tight control of booster roll.
But just one thing. So when we assume that the booster can hover. Won’t they be able to to change the rotation or position slightly by hovering above the Catcher Arms ?
question in some SpaceX broadcast video the commentator girl explained that in the last descent flights of the Falcon 9 they were not using the center of the target in order to take advantage of a little bit off center so that the gases from the suicidal combustion would not damage the drone ship's material excessively. If that is true, we would be talking about the fact that your count is not particularly in line with reality, since we would be talking about Space X being able to land the fan with 9 really where they wants.
Have they ever said that? I've heard it as a theory for a while but never anything officially confirmed. Based on the distribution we got it seems like they're always targeting the center, but hard to say 100%
There are so many moving parts and unavailable information on this I think that it wold be hard to make a better summation of this process, as far as we can know it from outside SpaceX. Going to be an interesting show.
Accuracy is relative landing on a drone ship vs a stationary target. The chopsticks have a fixed geolocation and the only movement that needs to be compensated for is the earths rotation. This would give a more consistent return trajectory. The drone ship has varying elevation and drift while the falcon is on it's return trajectory. The target interdiction in the drone's case becomes increasingly complex as the booster's altitude decreases. I can appreciate the conversational weight you've given to the landing's standard deviation though. A fair analog would be the landing accuracy of ground based targets/landing pads.
Gosh im way too excited for the catching attempt, someone take me to that moment now!!
This is a high-effort video. Love it.
@NASASpaceFlight EXCELLENT WORK excitement coming!
Cant wait for sunday!!!!
Btw, falcon 9 boosters don't hit the exact center of the droneship consistently BY DESIGN
SpaceX lands the booster on a slightly different zone by choice to not put a hole in the droneship's deck 😅
It's by design, if not, having boosters landing again and again dead center would put a nice hole in the droneships, the exhaust of a rocket motor is basically a cutting torch
Also add that falcon 9 can't hover, their thrust to weight ratio is too high, which is even more impressive cuz falcon 9 HAS to stick the landing in one shot, super heavy on the other hand CAN hover and adjust it's position between the chopsticks
Also don't worry, excitement is guaranteed!
It's gonna be a show even if superheavy doesn't get approved to land back on the launch platform, i honestly can't wait to see how the new heatshield performs 👀👀
The thing Superheavy has on the Falcon 9 is that it isn’t doing a hover slam. It’s actually gonna hover. So I would think Falcon 9 is more concerned with getting the hover slam timed than the accuracy on the target
Falcon boosters only have the grid fins and the Merlin suicide burn for final position control, the Super Heavy will have the grid fins and Raptors which can throttle down and basically have the booster hover if needed. Having it hover over the landing zone for too long would obviously wouldn't be good for stage zero but would give opportunity to adjust for it's position.
Also maybe the reaction control thrusters could be used to correct for any rotation forces imparted by the chopsticks or maybe not I've no idea how useful they would at sea level
I read somewhere that spacex deliberately doesn't land in the centre to reduce wear on the thermal protection
Great analysis DAS plus many insightful comments! I hope & guess SpaceX have it in hand. Go for it!
It's great the F9 realtime data feedback gives enough envelope to get the sims within tolerance for SH. Gonna be guaranteed exitement! Thanks for the analysis.
Great analysis! If I'm not mistaken, Falcon 9's booster landing position on the droneship is designed not to always land in the center of the droneship, so the droneship won't take the engine's fire damage every time on the exact same place.
One of several beautiful things about the process we see here is a relatively open discussion about potential problems. ("All of us is smarter than one of us") If you have attempted to solve difficult technical problems you know that this back-and-forth process is golden. Blue Origin seems to have decided that they don't need or want to have such interactions. The difference is obvious: slow, mistake-prone projects vs rapid convergence to a workable approach. Kinda looks like the the difference between the USSR and US in the first golden age of space exploration...
China, pay attention...
YES THEY CAN CATCH IT.
Regarding the roll of the booster needed to align the landing pin with the chopstick arm landing zone, they could increase the fudge factor by extending guide bars from the booster, above the landing pins, and placing a V-shaped trap on both arms, that would engage the extended guide bars some distance above the arm landing zone. Once the extended guide bars make contact with the V trap, the booster would be forced to rotate as the bars descend to the center of the V trap, thus aligning the landing pins with the chopstick arm landing zone.
I hope you guys keep running with this to see if there is a correlation between sea state and accuracy. Also, does one drone ship perform better than another? Does booster age have an effect? You really need the RTLS data. Nice show, thanks Das!
Thank you Das and NSF, love this videos👍👍
Great job with your analysis with the information you had. Impressive to say the least.
Great Analysis! And like Das said, "we'll see". 👍👍
Great analysis John. If this works, it'll be an awesome thing to see. I really hope they manage to do it.
I did learn something. thanks.
what would happen if superheavy misses the pins but lands on the gridfins, would they be able to hold it?
That's a topic of much discussion. Since it's not designed to do that I'm inclined to assume they would bend or break off, but maybe they're overbuilt enough to hold things together
Seriously, you can't even piss in the bowl from a foot away and you are going to pick on SpaceX about hitting the bullseye painted on a platform floating in the Gulf of Mexico with a 135.2 feet dart traveling at 22 miles per hour just before it hits the deck... 🤔
Loving it...😂
Great presentation Das. 👏👏👏👏👏
Its the variables that make this so damn exciting & frightening,
It's going be a great watch, whatever the outcome, roll on Sunday
I am extremely impressed with your work and persistence
How vulnerable will the booster be to strong winds during its landing manoeuvre? Will this affect the Super Heavies launch weather criteria potentially leading to a greater frequency of scrubs in comparison to Falcon 9?