Will Starship Fall Over?
Вставка
- Опубліковано 5 гру 2022
- Starship is a tall and thin vehicle that looks much different than the Apollo Lunar Module. What is the chance it will tip over when landing on the lunar surface?
@Eager_Space on Twitter
Triabolical_ on Reddit
/ eagernetwork
/ eager-space-1038430522... - Наука та технологія
Experienced KSP player here. Things that come to my mind (not sure hiw well they translate to the real world):
1. The landing legs help most if one leg directly points downhill. If the downhill direction is at 45° between two legs, the width of the base is significantly smaller.
2. Bouncyness: Landing legs are designed to take the impact of landing, which means they need some room to flex. During landing, *especially on slopes*, the lander will swing a bit back and forth before settling to its stable position. If during this motion, the COG is outside the ground area, it will tip over.
3. Torque: When the lander leans over, the downhill facing leg(s) need to bear a higher load. This causes them to give in more, amplifying the tipping. This problem is especially bad for tall landers.
That's why this starship render really scares me ...
Doctor K’s Engine Elixer: A more legitimate company than Established Titles!
Now Teller's Tachyon Tablets on the other hand is definitely one of those scam products, glad they are no longer sponsoring Eager
Seems pretty conclusive, other than the fact that the model does not account for margins in the propellant, or the leftover propellant in the header tanks. This is probably a good approximation.
Will HLS even have header tanks? I don't see a reason for them.
This and launchpad astronomy are becoming my favorite space-related content channels.
Thanks. I aim to please.
2250m/s is pretty tight for getting to NRHO. Starship might just manage it thanks to it's high TWR, but I'd imagine SpaceX will still include some extra margin on that, which will add a significant amount of weight thanks to the rocket equation. NASA budgets for ~2750m/s, depending on the exact profile.
Definitely; I went with the minimum because that would give the lowest propellant mass and therefore was the most conservative choice for this analysis.
Margin would definitely be a great idea. If you have a link for the NASA budget I'd love to read it.
@@EagerSpace
There's not much to read unfortunately, AFAIK they haven't justified their reasoning for any of their estimates.
2750m/s appears to be the most commonly used figure.
for example page 18 here: www.nasa.gov/sites/default/files/atoms/files/20191030-nac-heoc-smith-v3.pdf
and page 8 here: www.nasa.gov/sites/default/files/atoms/files/20181207-crusan-gateway-reduced-v4tagged.pdf
They used 2700m/s in a CLPS presentation and I think I even recall 2850m/s being thrown around back when Gateway was still DSG.
If I had to guess, I'd say that first braking into LLO and then landing as depicted in one of the graphics is less efficient than landing directly, and also that an inclination change will be needed after a stay on the surface unless the landing site is exactly at the pole.
And of course, they probably weren't expecting whatever lander they ended up with to be quite overpowered as Starship, which can easily manage a TWR well into the double digits, as opposed to the measly 2 that the Apollo ascent module had.
that bothered me from the first time I saw it... but I guess I have too many KSP memories... I will never forget the mission to save the mission which was supposed to save a tipped over lander but ended up tipping over...
Hey there, YT algo! Notice this comment. Use your powers of automatic sentiment analysis to see that I am happy.
Now start sending over more subscribers to technical deep-dive space channels such as Everyday Astronaut, CSI Starbase, Terran Space Academy, BPS Space, and others. This channel deserves a wider audience.
In the equation at 6:55, I see 9.8. I'm assuming that's for gravitational acceleration. If so, I think you might be using the figure for the wrong planet. If so, I think the fuel needed to reach orbit could be quite a bit lighter. I mean, you'd probably want some extra fuel for contingency. If I'm wrong, please let me know. If that's correct, it would be interesting to see how it affects the equations.
Apparently, 9.8 (or 'standard gravity') is the correct value here, because Isp is a weird bastard child of the metric and imperial systems.
The value that actually matters in the rocket equation is the exhaust velocity v(e) of your reaction engine, and v(e) = Isp * 9.8 en.wikipedia.org/wiki/Tsiolkovsky_rocket_equation
It would be interesting to take this stability model and deduce how much lateral velocity Starship could tolerate at touchdown on a level plane. I know, with computers this is realistically as close to zero as makes no difference, but still.
I considered doing that, but I was too lazy to do it. It would be fairly straightforward.
Not quite complete. But the rest of the calculation is more complex.
*It could tip over if it:*
(1) Contacts the surface a particular lean angle X ;
(2) AND if it contacts the surface with horizontal velocity above Y.
Yes, it's likely supposed to descend completely vertically.
But the above is relevant in case it does not.
Definitely true.
For #1, I included the Apollo 15 example which looks really tilted to me and yet Starship would be fine with that angle. Starship is probably a bit more immune to this because the landing pads are so much farther apart, but it's still possible.
For #2, I'm expecting that starship is probably going to autoland and that keeping the horizontal velocity low will not be a problem. Apollo was all under 1 m/s horizontal velocity and that was with hand control. Could still happen.
Worse case scenario: It contacts the surface at +X° and with Ym/s, then tips to the opposite -X° in rebound, then suddenly halts Y to 0m/s as a leg digs in, with +X being naturally inclined away from Y direction to halt the horizontal translation. Fuel slosh is not just the off-center weight, but horizontal momentum still traveling at Ym/s that must decelerate to 0 with an initial state already in the worst sloshed state, which is also not perpendicular to gravity, but swinging up the side.
Also geometry of the legs should assume they are at 45° (if 4 legs), not 90° to the slope direction, reducing the horizontal component of the leg to about 71%, and mounted further from the apex of the rocket body's radius as you rotate to 45°. On 9m body with legs extending out 8m from center, the mounting point of the leg at 45° is only 3.18m from the center of the body and the foot is only 5.66m from the center, barely over a meter wider than the body itself.
Fuel slosh would reduce the angle - so would lateral velocity at touchdown - so would landing on uneven ground.
Simplified 'back of the envelop; calculatins are useful only if you have a huge margin. If a simple static-force calculation says you can handle 20%, you can in fact handle 10%. So current Starship will be too unstable for exploration. HOWEVER Artemis is not exploration.
If the first mission to Shackelton crater has wide, heavy DISPOSABLE landing gear and the crew mark out level ground and subsequant Starships land on the level ground, a less stable design is OK.
If you would carefully place it like that with a crane, that is.
Landing it on that angle would not work out so great.
Thank you for this
I think starship's hull is nearer 4mm thick but that may change by the time they're ready to land it in the moon. It will also have landing engines higher up the hull and very possibly 9 raptor engines in the skirt.
Thank you for working this out and *demonstrating* the answer. There are too many assertions one way and another on various forums.
Btw, do you have a brother who does Greg's Airplanes and Automobiles? Your voices and cadences are very similar.
Nope, though I think you're not the first person to notice that.
severely underrated channel
What about propellant splashing? Because a lunar lander need to flip from a almost horizontal position to one horizontally relative to the local gravitational field, I mean would make that maneuver with only few tens of seconds before touching the ground, but leftover propellant would continue to splashing, obviously are some solutions like using smaller tanks that have only that propellants needed for final landing and lifting back, not header tanks but bottom tanks for extra stability.
Very relevant (:
How much horizontal velocity component at time of landing would be acceptable?
(Assuming that the target point is not 100% certain at time of approach and needs some adjustment just before touchdown. Unlike Falcon, where there's a defined landing surface, with HLS there's a non zero chance there might be some horizontal speed component at touchdown. Propellant sloshing as well.)
In this case you'd also have to look at fuel slosh.
Speaking of which, this gets me thinking about how you would do a long duration lunar mission with lunar starship. I doubt they can keep the tanks at the right temperature for weeks on end.
@@DemoR
You probably can't keep the heat out entirely, but if you can insulate the tanks enough you might be able to reach a level of boiloff where you can trade payload mass for longer durations. I.E , say you lose 1 tonne of propellant per day to boiloff. If you want to stay for 20 days, that means you'd need an extra 20 tonnes of propellant vs just touch and go, which means you have to trade off 20 tonnes of payload to make up that deficit. Just spitballing here of course.
I thought a bit about that but didn't do any real work on it.
The maximum horizontal velocity during apollo was less than a meter per second, and I'd expect that lunar starship would be better than that; I think they do much less than that with Falcon 9.
My *guess* is that 1 m/s would be fine.
@@DemoR There's obvious no convection to worry about and the only conduction will be through the landing legs, so pretty much nothing there.
So the only heating you get is solar, and with white paint that should be pretty minimal. Vaporizing the cryogenic fuels requires a lot of heat, so I think they will probably be fine there.
the reaction wheels should be able to handle this :)
How on Earth in Starship supposed to keep its fuels at cryogenic temperatures for the weeks it would take to complete a lunar mission?
How much fuel can Starship expect to have off-gassed by the time it needs to launch into lunar orbit?
Are they expecting to take an alternative energy source to keep these fuels refrigerated over those weeks?
Or are their tanks just absurdly well insulated in vacuum flasks to keep LOX cold for weeks in direct sun, with all the weight that implies?
An engineering analysis could be done to quantify this, but I haven't do it, so here are my thoughts.
The only heat gain on the moon or in orbit is from solar radiation. Lunar starship is white to reflect as much of the energy as possible, but some will heat up the structure.
When you put heat into cryogenic liquids, they don't heat up but rather some of the liquid turns into gas. That absorbs a lot of heat so that the rest of the liquid stays liquid. So the real question is how much of the propellants they will have left.
My understanding is that it won't be significant in this sort of scenario, but I haven't done any analysis. I do know that NASA has done some work to develop techniques to keep liquid hydrogen liquid for long flights, and that's much harder to do than methane or oxygen as the hydrogen temperatures are so much colder.
@@EagerSpace Can they spend some electricity to cool the propellants? How much solar panels or batteries will that take?
Compressibility (or maybe there are more appropriate terms) of the regolith is another factor I would guess.
It's not will it fall over; it's when it falls over,
"Then what will you do."
How much more chicken broth would it take if the legs sink 2 meters into the soil on one side?
Cool
'Lokhotron' 😂
You forgot about calculating the fuel that is used for backfilling ullage.
Starship cannot land and still stay standing safely at 20 or less degree angle. As soon as you open the cargo bay doors and start lowering cargo from the nose, you're putting weight outside that 20° angle.
Yeah, and lowing it, which makes it even more stable.
I'm a spherical viewer (and liker).
Do the same for 70m height Starship V3 please
Should be fairly simple given the information released.
yes if pushed by a giant
While we're discussing Starship and the Moon - could you do a video on Starship alternatives to SLS/Orion? Your ability to give hard figures and clearly spell things out is sorely needed for this. I've seen a proposal on reddit that a lightly loaded regular Starship with a crew compartment could go to NRHO and back *without needing refilling* in NRHO. Not needing a refill is necessary for this to be considered in the near- and mid-term by NASA.
The crew compartment for 4-6 can be based on the one for HLS, ergo NASA approved. A light load of cargo will be fine early in the program. Heavy cargo-only runs can be done separately. I hope to see this done at about the time of Artemis 4 so a crewed launch on the Starship isn't included. Crew would launch on a Dragon and board the refilled ship in LEO.
The return of the crew will require a bit of a kludge. A stripped down Dragon is stowed in the cargo bay at launch and only deployed when the Starship is closing in on Earth. The crew deploys in this and enters the atmosphere & splashes down conventionally. The ship lands autonomously. A stripped down Dragon needs no SuperDracos and only minimal Draco propellants, along with minimal ECLSS and a stub trunk - all just enough for a few hours. Carrying this all the way to the Moon and back is certainly a bad looking kludge but is the most straightforward way to make NASA happy about a crew-rated reentry. Even SpaceX may not have crew-rated it by the late 2020s, Elon wants a couple of hundred successes *in a row.* A good enough solution that saves us billions by replacing SLS/Orion ASAP is preferable to a best solution later on.
All this would be easier if this Starship could carry enough propellant to and fro that would allow propulsive deceleration to LEO. However, I'm informed on reddit that the delta-v doesn't work out. This LEO-to-LEO proposal is made repeatedly, though, so it will be great if you can definitively disprove it, put a nail in its coffin.
Some guy on reddit has given answers to parts of this. Some guy named Triabolical.😉
I'm sure this looks familiar. A full-blown video will be so, so much better.
If it helps, I just subscribed.
It's on my list of topics to explore. It seems fairly simple to do it one-time, but the hard part IMO looks to be figuring out reuse for the lander; getting fuel out to NRHO to either do another landing mission or come back to LEO looks challenging.
Not sure when it will show up. I work on videos until I get bored, stuck, or aren't sure what my opinion is, and then I'll switch to something else. Or - often - I'll be working on one video and it turns out I need to do another one first; this video came out of a video on using starship as a habitat.
@@EagerSpace "It seems fairly simple to do it one-time". I'm afraid I don't see what you mean. The Dragon to LEO is reusable, the Starship for the round trip is reusable (it lands itself autonomously after the crew deploys in the capsule) and the stripped down Dragon can be reused. (It will of course have an upgraded heat shield.)
Yup, filling in space (LEO or NRHO) will be a hard nut to crack. Doing it just in LEO will be hard, and Artemis can't go forward without it. And the chain of tanker flights to get a significant amount of propellant to NRHO will need Musk's vision for Starship's high flight cadence to make it work. That's why I use the kludge of a stripped-down Dragon and no NRHO refill - I see it as an interim solution until Starship is human-rated or NRHO refilling is perfected.
Thank you for engaging. I certainly understand the need to be motivated when one is doing a project voluntarily, I've been there. But Christmas is coming... :) And besides me there are so, so many people looking for the LEO-to-LEO answer. You, sir, have a moral obligation to all of humankind. :D
@@donjones4719
When I said "fairly easy" I was referring to the artemis architecture where you go LEO => surface => NRHO and then leave it there.
I will take a run at it, and we'll see what pops out. Refueling probably complicates the number of approaches.
And I suspect there's an approach lurking out there that involves leaving the starship on the moon; I'm not sure taking 50 tons to the moon and taking a fair bit of it back makes a lot of sense. Or maybe there are just cargo variants that carry more.
> You, sir, have a moral obligation to all of humankind. :D
LOL. I shall therefore endeavour not to fail in this quest.
@@EagerSpace Multiple thumbs up. And don't worry about refilling scenarios for my sake. We have to take it on faith the initial LEO depot filling->HLS ship can be done, per NASA & SpaceX & their current HLS plan. It follows that my Starship can get its LEO filling the same way and then need no others. Elaborating your video to NRHO fillings would be gravy, don't let it bog you down if it becomes burdensome.
Instructions unclear. Where should I put my 135 CCs of chicken broth? I tried injecting the quail eggs but that didn't work... Regardless, great video!
Proper execution of the instruction is left as an exercise to the viewer...
Interesting calculations. There are a few more factors that could significantly contribute.
First, the terrain tilt may not be towards one leg, but between two legs. The deviation would then be closer to the half-side of the landing legs square, as opposed to half the diagonal. This significantly reduces the stability.
Second, the terrain may depress under one or more of the landing legs. It doesn't need to be much to throw the thing off-balance - considering that landing on tilted ground puts a lot more weight on the lower legs than on the upper ones, and Starship weights quite a lot.
Third, horizontal speed massively changes the max tilt angle. Starship would be particularly vulnerable to sideway speed, since its high mass would have high inertia. If the sideways speed is towards the lower end, the pressure on the lower legs would make them even more likely to sink in. I don't think this factor would be much of a deal, since computers would be in control, unlike in Apollo, and they would likely negate the horizontal speed.
In all, Starship would be crazy efficient if there is a landing pad prepared. But the first landing would likely have much larger legs and probably ballast in the lower parts, and the rovers it carries can prepare landing pads for the next landers. I'm sure SpaceX would do their homework so the lander does not fall over.
Doesn't matter if it tilts towards one leg or two, it's the net that counts. They already know the terrain doesn't compress very easily but it's of course not impossible. They'll kill the horizontal speed much the way Apollo did and likely try to come in as straight up and down as possible, they won't be landing under manual control. Lastly landing legs can be made to compensate for a lot including self leveling.
It’s not going to the moon
can they not map the height at each leg as it descends and mechanically adjust the legs height for standing position, using actuators to either telescopically adjusting the length or using pistons the angle, a bit how a four legged robots cybernetics will autobalance, Personally I think as long as the landing spot has been vetted in advance by orbital drone this shouldn't be a big problem as a suitable site can be selected, I am far more worried about regolith kicking up when taking off and breaking a thin pipe or the engine bell, or even just blocking some of the outlets with regolith.
Yes, the landing legs can self-level to a degree. Regolith will be moving away from the ship when the engines are firing, not towards it.
@@TheEvilmooseofdoom this actually happened with the martyte concrete platforms while they were testing SN8, they have covers now, but still...