Debug: Database connection successful
You are not logged in.
For Louis re #24
Thank you for the link to the Occupy Mars video. I watched it when the count of views was 10,656, and after finishing the count was up to 13,230.
The video is well done, with plenty of "live" video and animation. The Mars Society shows up in minute 40:25, with mention of its two design competitions.
For GW Johnson re #25
Thank you for your careful attention to the "real numbers" that will govern what actually happens!
If all the scenarios laid out in the Occupy Mars video come to pass, it will be because the physics allows them, and the human designers paid attention.
(th)_
Offline
Like button can go here
Jixuan & Sebastian report Elon Musk has reaffirmed 2020 for the Mars cargo flight! A big yay from me!! It will be incredibly tight but on the plus side, no human will be used in the making of this cargo mission...so perhaps Space X can afford to gamble...
https://www.youtube.com/watch?v=1J3OV27w-6M
My issue would be - "What will be on board?"
It could carry some feedstock for propellant production I guess...not a very productive use of cargo.
It could maybe carry some ATK solar equipment.
But we have heard nothing to suggest that Space X has developed anything like a Mars-proof automated propellant production plant or autonomous robot mining rovers.
I guess you could pack in medical supplies, water, a couple of methane-oxygen electricity generators and dried or tinned foods that will last till 2026. You could probably commission an experimental hydroponic food hab on the Antarctic station model, which could be packed on board. Also, I supposed Tesla batteries would be useful,so they could go on board.
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
Someone, perhaps Ryan MacDonald, mentioned that the Starship/SuperHeavy though about the same size as Saturn V will have double the payload to LEO...so while the physical constraints may not change, our ability to work within them certainly can over time.
Louis:
You find all sorts of stuff said on the internet by all sorts of people, pushing all sorts of agendas. You find very few on the internet who actually understand how to run the numbers. The numbers don't lie, as long as you use the right ones in your equations. There is a garbage-in/garbage-out problem that many run afoul of. I usually don't, because I am one of those rare ones who actually did this stuff for a living. I have been long-retired from that, but the physics has not changed, and the technology has not changed very much.
If you look at the evaluations I did (posted on "exrocketman" for all to read) for the 2020 and 2019 versions of the Starship/Superheavy design concept, you will see that the 2020 version is less capable than what they thought in 2019. That is precisely because the inert mass is nearer 120 tons than 85 tons, despite the growth in propellant loadout from 1100 tons to 1200 tons. With a similar 100 ton increase in Superheavy propellant, I might add.
Now, what I showed for the 2019 version was no capability to enter Mars orbit. There was simply not the mass ratio in the design to support an orbit entry burn in addition to a Hohmann transfer and the powered landing. It had to reduce payload to reduce transit time with a faster trajectory.
There are those who point at entering an elongated capture orbit with repeated perigee aerobraking to capture at Mars, but that notion has two very critical downsides: (1) it has never been successfully done at Mars before, and (2) the results are erratically-factor-of-2 variable because the upper atmosphere density is factor-of-2 erratically variable at Mars.
This is aggravated by faster trajectories, because the bigger your initial approach velocity, the more likely you are to bounce off into deep space instead of capturing at all. There is no backup propulsion capability available, the mass ratio does not support that.
Now as I already said, the 2020 version is less capable. It carries less payload than the 2019 version. 2019 could not execute an orbit entry burn and carry any payload. 2020 most certainly cannot. The numbers say that without a doubt. And the enormous risks of attempting repeated-aerobraking capture into a variable atmosphere at extremely-marginal half-again-to-twice-escape-speed conditions make that notion wishful thinking at best, and most likely arrant nonsense.
I'm sorry, but the numbers do not lie. Not as long as you put the good stuff through the equations.
And, I would point out that the Mars entry simulation Spacex has had on its website shows a trajectory and an entry velocity incompatible with repeat-pass aerobraking. It is compatible with direct entry from the interplanetary trajectory, one just a tad faster than Hohmann min-energy transfer. The entry speed they show is 7.5 km/s max. Mars escape is 5.03 km/s. Entry from low Mars orbit would be 3.6-3.7 km/s.
GW
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
I know there has been talk of going to orbit instead of the surface with Starship. I am personally very doubtful of that, because the mass ratios as we currently understand them do not support that, plus how do you land and refill? Some point at repeat-pass aerobraking to reduce or eliminate the propulsion needed to get into orbit. I took a good hard look at that. The results are posted over at "exrocketman" today. Because of factor-2 upper atmospheric density variation, you cannot eliminate the need for a fair fraction of the propulsion you would need for rocket-burn into low Mars orbit. And you have to tailor and execute the delta vee "on the fly" as you exit from your first braking pass. It is doable, but it is also very demanding. And it cannot save you as much propellant as you would like.
GW
GW Johnson
McGregor, Texas
"There is nothing as expensive as a dead crew, especially one dead from a bad management decision"
Offline
Like button can go here
I don't have the technical knowledge to comment on the feasibility of aerobraking into Mars orbit but I found this from Blake MW on Reddit helpful:
"It's useful to consider Newton's Impact Depth theory, essentially when something going at hypersonic speed runs into a medium it'll run out of momentum once it has "swept up" it's own mass in whatever medium it is colliding with (whether that's atmosphere or water or rock, it doesn't really matter at hypersonic speeds).
I'll treat the BFS cross section as a 9x45m rectangle or 360m2.
The density of martian atmosphere near the surface is 0.020kg/m3 - and the BFS in the animation does almost skim the surface (altitude of 6km).
The BFS will sweep up 360m2 x 0.02kg/m3 = 7.2kg/m of forward motion.
I'll take the BFS total mass as 150000kg
The BFS will need to sweep through 150000kg / 7.2kg/m = 20.8km until it has swept up its own mass in atmosphere.
Now obviously this is a very crude approximation and in particular it completely breaks down at super/subsonic speeds when air can easily get out of the way (that's why about 600m/s has to be eliminated by other means), but if we consider the BFS coming in horizontally across the surface (skimming the surface) with its belly forward, it actually runs into plenty enough atmospheric mass to eliminate most of its momentum.
Considering that the BFS is entering at up to 7.8km/s, it would, if it could get close to the surface while still traveling that fast, be running into around 56t of mass per second! In fact, the g-forces can be quite severe when aerobraking at Mars. SpaceX appears to handle this by slowing down in the higher atmosphere, using negative lift to follow the curvature of Mars, then rolling upright in the lower atmosphere and using positive lift to buy more time for running into air before hitting the ground. Watch the true physics simulation to see what I mean by using negative lift at high altitude to follow the curvature and positive lift at low altitude to gain altitude.
It's important to not mistake 0.6% of Earth's atmospheric pressure as being negligible. Earth's atmosphere is not negligible only because it has 1.1% the pressure as Venus's atmosphere. Mars has plenty of atmosphere for many purposes."
I know there has been talk of going to orbit instead of the surface with Starship. I am personally very doubtful of that, because the mass ratios as we currently understand them do not support that, plus how do you land and refill? Some point at repeat-pass aerobraking to reduce or eliminate the propulsion needed to get into orbit. I took a good hard look at that. The results are posted over at "exrocketman" today. Because of factor-2 upper atmospheric density variation, you cannot eliminate the need for a fair fraction of the propulsion you would need for rocket-burn into low Mars orbit. And you have to tailor and execute the delta vee "on the fly" as you exit from your first braking pass. It is doable, but it is also very demanding. And it cannot save you as much propellant as you would like.
GW
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
That's not the method I use for entry aerodynamics and flight mechanics. I do it a different way.
What the Spacex website claims is entry at speeds up to 7.5 km/s, not 7.8 km/s. But once you are that fast, it makes little difference. It's about 7.9-8.0 km/s coming back from LEO, which makes those entries rather comparable, until you decelerate out of the hypersonics at about Mach 3.
The two entry simulations (Earth and Mars) for Starship are rarely posted on the Spacex website at the same time. The entry sequence for Mars is fundamentally different from the sequence for Earth, which is traceable to the fact that Mars's atmosphere resembles Earth's upper atmosphere, if you compare Mars's surface to Earthly conditions at about 105,000 feet (33 km).
On Mars, they come out of hypersonics at about Mach 3 at only about 5 km, just like I have been telling everyone for some years now. From there, they have to pull up into a near-tail slide maneuver at about 10 km and around Mach 2. That puts them tail first with the rockets "on", still supersonic at 5 km again for the final touchdown. It's a real nail-biter of a descent sequence. And I'm betting they have to use thrust to augment the lift in order to make that pull-up. They don't think so, not yet. But they eventually will, I predict.
On Earth they come out of hypersonics way up around 40-45 km altitude, and as the trajectory bends downward, they pitch up from 60 degrees during entry to a full broadside 90 degrees. They then fall dead-broadside like that, from supersonic in the thin air above 33 km, to well-subsonic (supposedly 68 m/s) in the thick air below 5 km. Then they pitch to tail-first, lighting the engines, and touch down. That's the "belly-flop" or "skydiver" descent, which is ABSOLUTELY IRRELEVANT to descent on Mars. The two planets are just utterly different in their descent requirements.
GW
GW Johnson
McGregor, Texas
"There is nothing as expensive as a dead crew, especially one dead from a bad management decision"
Offline
Like button can go here
I don't think there's any dispute about the different requirements of Mars and Earth entry. Paul Wooster seems to accept that fully.
That's not the method I use for entry aerodynamics and flight mechanics. I do it a different way.
What the Spacex website claims is entry at speeds up to 7.5 km/s, not 7.8 km/s. But once you are that fast, it makes little difference. It's about 7.9-8.0 km/s coming back from LEO, which makes those entries rather comparable, until you decelerate out of the hypersonics at about Mach 3.
The two entry simulations (Earth and Mars) for Starship are rarely posted on the Spacex website at the same time. The entry sequence for Mars is fundamentally different from the sequence for Earth, which is traceable to the fact that Mars's atmosphere resembles Earth's upper atmosphere, if you compare Mars's surface to Earthly conditions at about 105,000 feet (33 km).
On Mars, they come out of hypersonics at about Mach 3 at only about 5 km, just like I have been telling everyone for some years now. From there, they have to pull up into a near-tail slide maneuver at about 10 km and around Mach 2. That puts them tail first with the rockets "on", still supersonic at 5 km again for the final touchdown. It's a real nail-biter of a descent sequence. And I'm betting they have to use thrust to augment the lift in order to make that pull-up. They don't think so, not yet. But they eventually will, I predict.
On Earth they come out of hypersonics way up around 40-45 km altitude, and as the trajectory bends downward, they pitch up from 60 degrees during entry to a full broadside 90 degrees. They then fall dead-broadside like that, from supersonic in the thin air above 33 km, to well-subsonic (supposedly 68 m/s) in the thick air below 5 km. Then they pitch to tail-first, lighting the engines, and touch down. That's the "belly-flop" or "skydiver" descent, which is ABSOLUTELY IRRELEVANT to descent on Mars. The two planets are just utterly different in their descent requirements.
GW
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
Boston Dynamics robot dog spotted (ha-ha) at Boca Chica -
https://www.youtube.com/watch?v=ZXxRXJ7o1N8
A collaboration with Boston Dynamics could be a very interesting one.
Maybe they could use a robot spider-dog to climb the return Starship on suction pads to undertake a close visual inspection of all parts.
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
For Louis re #33
While your phrasing suggests (to me at least) you were offering some light hearted humor, I think your idea is quite interesting.
I asked Google from some coaching on the question of magnetic response of stainless steels .... Apparently the alloy mixture that gives stainless steel its desirable properties works against magnetic susceptibility, but not entirely.
Beyond that however, is the simple proposition that a robot that can walk on Earth can surely navigate its way around a space ship using little puffs of gas as needed, while orientation can be achieved using gyroscopes.
In short, a version of the robot dog adapted for space flight would seem a likely member of an expedition crew.
Thanks for spotting this development!
(th)
Offline
Like button can go here
Yes (th).
Louis, you made me look. Now I have to post before breakfast.
I visualized something this morning.
Could Spot be sent to the Moon with a variation of Lunar Starship?
A one way trip, with no LEO propellant refilling?
Tricks would be needed, and even then, I don't know if it would have enough propellant.
But it is said that Starship cannot land on the Moon very well, or perhaps at all with raptor engines.
Depictions seem to show what could be clusters of engines further up on the body of the ship. Propellant type not known.
Some people however think they are lights for landing in the dark. It seems sure that SpaceX will have smaller engines somewhere on the ship to land on the Moon.
So, what I think, is that you split Deep Space Starship into two stages. The small engines go with the upper part to the Moon.
Super Heavy gives the first boost. Then the lower part of Starship gives another boost, up to where the atmosphere is very tenuous, but still significant. The lower part of Starship drops off and de-orbits to burn up and splash in the ocean presumably.
The upper part uses the small engines to go to orbit and then to the Moon.
It might have deployable legs, and a dog house for spot at it's base. Spot will have to have a charging station, and a warm place to hide. Spot may be durable enough to go out in the mornings and afternoons, with a full charge.
Then you have a semi-autonomous remote Lunar explorer, that has some telepresence capabilities at a location desired on the Moon.
Of course NASA/Government could permit a nuclear power source of some type, or it could have deployable solar panels.
The device may be able to have some of the sample testing equipment that the Mars probes have.
Spot would have a laser perhaps on it, and they could put a hand on it (They have those already), and a backpack.
I am thinking that if they could do it, this would be a product they could sell.
------
But Spot is very likely in Texas for testing at this point.
Done.
I don't want to start a new post, so, I ill amend this one.
I have felt that a rickshaw wagon could be useful for a astronaut on other worlds. Carry more consumables, and maybe have a solar power roof.
I am thinking whereas I previously described Spot as scouting a particular location, I think she should have a wagon-kennel.
That way Spot could depart from a lander, and then just head out. The wagon would need solar power, and batteries. Spot could shelter in her kennel when it was too hot or cold.
But the "Payload" may be more limited, so that you might have to pick and choose the instrumentation for the mission.
In this case Spot would be a scout.
The wagon would also help stabilize spot on some terrains. Spot would grab the handle with her hand and then away she goes.
Also speaking of the small landing engines, I would think there would not be a need for redundancy, so reduce the number of thrusters, if it would still work.
And as for the last stage-lander, it would not have to be the standard length. Whatever might work best.
So, then telepresence scouting could begin on the Moon. Real people on Earth could have paying jobs working with it.
Who knows, maybe rich people could pay to have their own Spot on the Moon
Done
Last edited by Void (2020-06-19 12:56:33)
End
Offline
Like button can go here
300-series austenitic stainless steels are non-magnetic. You can explore this for yourself with what's in your kitchen. Try a kitchen magnet on the blade of a stainless steel dinner knife from your cutlery drawer. It does not attract at all.
Meanwhile, I see Spacex has moved from 301 stainless to 304L stainless in its prototypes. My old copy of Mil Handbook 5 only lists 301, and indicates all them are similar, because they really are. But the success is in the details, where they differ. Most stainless steel piping and plumbing fittings are made of 304 for a reason. Actually several. 304 will routinely go quite a ways hot and also cryogenically cold, for one. All of these are cold-work hardenable, but NOT heat treatable.
If the assembly is to be welded, the version of 304 to use is very most definitely 304L, not baseline 304. If you weld on 304, it will crack, usually upon cooldown, or not very long after. 304L has had its composition slightly adjusted to prevent that. It has served us very well for over half a century now.
Typically, about the highest temperature you want your 300-series stainless to reach is in the 1000-1200 F class. It will corrosion-scale if you let it get too hot. 304/304L starts doing that at about 1200 F. 316/316L and 347 will go hotter (nearer 1600 F), and 309/310 hotter still (about 1900 F) without scaling, but none of those are suitable to go cryogenic, too. It is 304/304L that gets used to make cryogenic tanks. Even so, all of these have about the same hot strength characteristics. They are down to around 5 ksi tensile at 1000-1200F, and really soft "butter" any hotter than that.
There are martensitic stainlesses like 4130 which are less corrosion resistant, and heat-treatable. They are also magnetic. But they also peak out near 1000-1200 F, and they don't go cryogenically cold. If you stay under the annealing temperature (down around 800-900 F), the heat treating survives, and you can have very high room temperature tensile strength. Usually, you do NOT want to go to max strength, because the elongation-to-failure gets too small (it behaves as a very brittle material, like glass), shattering at the slightest provocation.
There are some other alloy steels which can go hotter with better strength than stainless, but they are subject to severe scaling and corrosion, and they do not go cryogenically cold. They are magnetic. 17-7PH is my favorite among these.
All of which merely goes to say "you can't have everything". If your cryogen tank must also survive entry heating, then your magnetic space boots will not work on it.
GW
Last edited by GW Johnson (2020-06-19 08:17:27)
GW Johnson
McGregor, Texas
"There is nothing as expensive as a dead crew, especially one dead from a bad management decision"
Offline
Like button can go here
Even when BFR and starship is flying to mars the customers are not private citizen but are paid solders of the funding company or government sponsoring the flight.
To settle means one must not only be able to pay the trip costs but to also bypass the regulations which could prevent the launch to places off world.
Offline
Like button can go here
Actually I did say suction pads, as I had vaguely recalled the steel was non-magnetic. However, thinking about suction pads, don't know if they would work on Mars...can you have reverse suction pads?
300-series austenitic stainless steels are non-magnetic. You can explore this for yourself with what's in your kitchen. Try a kitchen magnet on the blade of a stainless steel dinner knife from your cutlery drawer. It does not attract at all.
Meanwhile, I see Spacex has moved from 301 stainless to 304L stainless in its prototypes. My old copy of Mil Handbook 5 only lists 301, and indicates all them are similar, because they really are. But the success is in the details, where they differ. Most stainless steel piping and plumbing fittings are made of 304 for a reason. Actually several. 304 will routinely go quite a ways hot and also cryogenically cold, for one. All of these are cold-work hardenable, but NOT heat treatable.
If the assembly is to be welded, the version of 304 to use is very most definitely 304L, not baseline 304. If you weld on 304, it will crack, usually upon cooldown, or not very long after. 304L has had its composition slightly adjusted to prevent that. It has served us very well for over half a century now.
Typically, about the highest temperature you want your 300-series stainless to reach is in the 1000-1200 F class. It will corrosion-scale if you let it get too hot. 304/304L starts doing that at about 1200 F. 316/316L and 347 will go hotter (nearer 1600 F), and 309/310 hotter still (about 1900 F) without scaling, but none of those are suitable to go cryogenic, too. It is 304/304L that gets used to make cryogenic tanks. Even so, all of these have about the same hot strength characteristics. They are down to around 5 ksi tensile at 1000-1200F, and really soft "butter" any hotter than that.
There are martensitic stainlesses like 4130 which are less corrosion resistant, and heat-treatable. They are also magnetic. But they also peak out near 1000-1200 F, and they don't go cryogenically cold. If you stay under the annealing temperature (down around 800-900 F), the heat treating survives, and you can have very high room temperature tensile strength. Usually, you do NOT want to go to max strength, because the elongation-to-failure gets too small (it behaves as a very brittle material, like glass), shattering at the slightest provocation.
There are some other alloy steels which can go hotter with better strength than stainless, but they are subject to severe scaling and corrosion, and they do not go cryogenically cold. They are magnetic. 17-7PH is my favorite among these.
All of which merely goes to say "you can't have everything". If your cryogen tank must also survive entry heating, then your magnetic space boots will not work on it.
GW
Let's Go to Mars...Google on: Fast Track to Mars blogspot.com
Offline
Like button can go here
For Louis re #38
Yes, you did say suction pads.
Your suggestion of suction pads is understandable. I can easily imagine how it might play a part in designs for inspection of vehicles and structures.
I was thinking more about the long term. Magnetic "walking"/navigation on the exterior of space structures would be advantageous, if engineers can provide for the capability. As you have probably noted, ISS external space 'walks" appear to be done almost entirely by hand, supplemented by safety lines that have to be clipped and unclipped very few meters of travel.
GW Johnson has already provided part of the long term answer. He pointed out that certain versions of stainless steel have a small magnetic component, but these are not preferred for environments where materials can rust. However, for a vehicle that will spend hundreds of years in open space, it is likely to be found acceptable to include a partial magnetic susceptibility in the specification, and magnetic "walking" is a much better solution than a suction system.
Magnetism operates without constant input of energy, if permanent magnets are employed in the "feet.
On the other hand, an efficient way of creating a vacuum when space itself is one of the most effective vacuums known to humans will prove challenging for the engineers hired to find a way to accomplish it. You'll place your foot against the surface of the vehicle while a perfect vacuum surrounds the foot, and then (presumably) try to evacuate gas from under the foot.
Another option is chemical adhesion. I'm unsure of how well that method works on stainless steel (or aluminum) in free space, but it certainly is worth investigating.
(th)
Last edited by tahanson43206 (2020-06-19 17:31:14)
Offline
Like button can go here
Type 304L is a very low carbon Stainess steel. I believe something like 0.03% Carbon. This makes the steel far more ductile and less prone to cracking under loads.
Last edited by Oldfart1939 (2020-06-19 19:25:34)
Offline
Like button can go here
Precipitation hardening alloys are subject to overaging at high temperatures. This is when the defect pinning deposits within the grains get larger and migrate towards the grain boundaries by diffusion, becoming less effective in hardening the material. The result is a reduction in strength and an increase in deformation. I wouldn't use them at extreme temperatures, unless someone has come up with a new type that doesn't suffer from this issue.
Offline
Like button can go here
For elderflower re Post #41
Design and selection of materials for rocket bodies is what I ** think ** the two of you (with GW Johnson) are discussing. It appears to me that at the moment, SpaceX concentration is on finding the best material to allow for launch and then the challenge of re-entry.
However, there is a use case that Louis brought up with his posting about the robot dog visiting the SpaceX Texas site. That use case is the structure designed for permanent use in deep space. Over time, I expect the great majority of long lasting vehicles in service in the Human Solar System will be those which move from orbit to orbit.
In that context, a modest magnetic susceptibility would be useful for periodic maintenance and inspections by human workers but primarily by robotic assistants.
GW Johnson recently hinted at the possibility there might be one or more alloys of steel that might be at risk of chemical deterioration in an atmosphere, but which would have superior qualities for permanent space structures.
Is there a material you would recommend (along these lines) that would combine strength, flexibility to endure heating cycles due to solar energy presence or absence, and modest magnetic susceptibility, along with other qualities you may consider best for this use case?
(th)
Last edited by tahanson43206 (2020-06-20 05:52:10)
Offline
Like button can go here
Both Elderflower and Oldfart1939 probably know more about metallurgy than I do. They might well be much better sources than me for materials to serve in space for a century or so. My experiences were mostly rocket and ramjet motor cases for one-shot missiles.
Suction means vacuum, or at least, relative vacuum. For a suction cup to stick to a surface, the absolute pressure inside the cup must be less than the absolute pressure outside the cup, and there can be no leakage into the cup. The suction force is the pressure difference across the cup times the flat reference area of the cup.
That being said, if the absolute pressure outside the cup is zero (vacuum of space), the absolute pressure inside the cup CANNOT be any lower than zero, zeroing the pressure difference across the cup, so there can be NO SUCTION FORCE using such a device on surfaces, while out in space.
And, the notion of "reverse suction pads" is a non sequitur. There is no such thing as "reverse suction". Precisely because there is no such thing as a negative absolute pressure.
As to materials selection for this or that application, it is the application that drives this, not anything intrinsic to the material candidates themselves. I think both Oldfart1939 and Elderflower would agree with that statement.
For a cryogen tank, you want a material that is both strong, and not brittle, when cold. Many materials satisfy that, including stainlesses, that lithium-aluminum alloy stages are made of, and even composites if appropriately lined. But, if you want that cryogen tank to survive many chilling cycles, that eliminates the aluminum, and it likely eliminates any practical liner in your composite. Of the stainlesses, only 304/304L has the right characteristics for many cycles.
If your cryogen tank must also survive exposure to entry heating, you cannot let the temperature get too high, or else you must cover up its surface with a heat shield. As it turns out, 304/304L will take a fair amount of heating, and still survive many cycles. It will do this job better than 301 or any of the other 300-series stainlesses, so guess why Spacex changed their material from 301 to 304L?
There is another fundamental aspect to this cryogen-tank-surviving-entry-heating scenario. You CANNOT have any liquid cryogen in the tank when it is exposed to entry heating. That liquid will very quickly boil, over-pressuring the tank. It will burst. Or you have to have a self-controlling venting means, of large flow capacity, which is heavy and expensive. Plus you are losing precious propellant.
So I give you ONE GUESS as to why Musk's 2016, 2017, and 2018 presentations showed small tanks nested inside the big ones! Those he referred to as "header tanks". You do that to isolate the cryogens for the next burn from the hot wall of the outer tank during entry. If you guessed that, you win the proverbial cigar. If not, you need to educate yourself further about what ALL is really going on during spaceflight.
And don't kid yourself, there must be cryogens for the next burn in the Starship of the Starship/Superheavy design. The bulk of the Starship propellant loadout is expended boosting to orbit, or if refilled on-orbit, departing from LEO to the moon or beyond. The remainder is in the header tanks.
In the case of LEO operation, there are two upcoming burns: (1) deorbit, and (2) landing.
In the case of outside-LEO operations, there are at least two upcoming burns: (1) course correction, (2) landing (moon or Mars), (3) takeoff from the moon, (4) course corrrection returning from the moon, and (5) Earth landing returning from the moon.
If refilled on Mars, the bulk of the propellant loadout is expended escaping from Mars onto the interplanetary trajectory home. There are two upcoming burns: (1) course correction, and (2) landing on Earth.
In all these cases, the empty outer tank also serves another critical function. It helps limit the conduction of heat into the nested header tank, precisely because it is empty, and (2) it acts to completely shield the nested header tank from solar radiation heating. This is necessary for maintaining the cryogens as cryogens for months-long durations. It reduces the size and weight of the cryocooler equipment you MUST have to do these missions with cryogens at all.
I know this is complicated. But that's just the way it is. We must deal with that.
GW
GW Johnson
McGregor, Texas
"There is nothing as expensive as a dead crew, especially one dead from a bad management decision"
Offline
Like button can go here