Jump to content

Artemis Discussion Thread


Nightside

Recommended Posts

3 hours ago, Exoscientist said:

Thanks for that. I was irritated by the approach SpaceX was taking towards the Starship in that SpaceX was ignoring basic principles in spaceflight. The main point was that for high delta-v missions such as lander missions to the Moon or Mars, these are done most efficiently by adding additional stages. But instead SpaceX wants to keep the SuperHeavy/Starship  as two-stages and instead do multiple refuelings. Another irritation was that SpaceX was ignoring the basic principle that you want to minimize upper stage dry mass as much as you can.

The SpaceX approach is to make a 100%, rapidly reusable rocket that is capable of landing back on Earth, and at Mars.

THAT IS THE APPROACH.

They are not making your pet rocket to do cislunar flags and footprints. They take the path WRT NASA that they always take—if a NASA goal is useful to them, they'll let NASA offset some of the cost of getting that tech they want to the right TRL. Cislunar? Sure a few billion bucks to offset the costs of nailing refilling, and life support on Starship. This is exactly the same as getting customers to pay for an expendable launch—then experimenting with landing the already 100% paid for booster. Failure to understand this is a failure to understand why they do what they do. You can not be onboard with colonizing Mars (I'm not, for instance)—but THEY ARE. View everything through that lens.

3 hours ago, Exoscientist said:

However, in making the Starship reusable, a whopping 80 tons, and with that much being subtracted from LEO payload, was added to the Starship being made reusable, going from a possible 40 ton expendable dry mass to a 120 ton reusable dry mass.

See above. 100% rapid reuse.

3 hours ago, Exoscientist said:

 Why is this approach taken? Because SpaceX has this fixation that the Starship must be the be-all-end-all for ALL of spaceflight. Then to get the cost per kilo they want, they need the Starship to be reusable. Then any extra mass added to dry mass is acceptable, no matter how much it is.

Yes. It doesn't make a difference at all. It's cheaper to by a "stateroom" first class seat to the other side of the world than it is to buy a Gulfstream capable of that same flight, then destroying it upon landing, buying a new one for the return flight. Reuse is everything, throwing rockets away is STUPID if you can figure out how not to.

3 hours ago, Exoscientist said:

 But a simple cost analysis shows this is a fallacious approach. A SpaceX exec estimated “10ish” refueling missions for the Artemis lander missions using the Starship HLS. But Elon estimated an aspirational $10 million launch cost for a reusable SH/SS. But this would be $100 million for 10 reusable launches. But a single expendable launch would only be $90 million by which because of the higher payload SH/SS could do single launch missions both to the Moon and Mars. No refueling missions required, and not even the SLS being required. 

These 2 are not the same. The single-stack expendable lands a smaller vehicle on the Moon vs a ~3500 sqft house. It also doesn't move the ball for them.

Note that NASA could have made an upper limit on habitable volume for Artemis. They could have asked for a crew vehicle of no more than 300 m3, where the airlock was no more than X meters off the lunar surface. They didn't (and it's not like they didn't expect SpaceX to bid Starship).

Edited by tater
Link to comment
Share on other sites

But Starship isn’t exactly reliable. It fails almost every flight, and frankly I don’t even know why they’re shooting for orbit, considering it could barely go to sub orbit. And Starship is taking so long to fix, yet it’s advertised as the future of space travel. Also, let’s talk about Elon himself. You really want our future to belong to a Reddit bro? He’s openly anti-Semitic, has 10 children among three women, half of whom won’t speak to him. But no, he can run a colony. He can’t even keep his own family together, and somehow you expect to run a city on another planet. People just listen to anyone with money.

Link to comment
Share on other sites

38 minutes ago, Kerbalsaurus said:

But Starship isn’t exactly reliable. It fails almost every flight, and frankly I don’t even know why they’re shooting for orbit, considering it could barely go to sub orbit

This is completely false. The IFT flight hit SECO 146 m/s shy of a Crew Dragon SECO. Intentionally shy of orbital. Had they burned 4s more (or a slightly different direction), orbital.

38 minutes ago, Kerbalsaurus said:

And Starship is taking so long to fix, yet it’s advertised as the future of space travel.

Faster turn around than early F9 flights at this point, you're clueless in addition to the innumeracy.

 

Edited by tater
Link to comment
Share on other sites

6 minutes ago, tater said:

Faster turn around than early F9 flights at this point, you're clueless in addition to the innumeracy.

But like… Starship has only had 3 orbital flights in the span of 11 months. Not as rapid as the F9, which has flown 27 flights since the new year.

Link to comment
Share on other sites

1 hour ago, Kerbalsaurus said:

But Starship isn’t exactly reliable. It fails almost every flight, and frankly I don’t even know why they’re shooting for orbit, considering it could barely go to sub orbit. And Starship is taking so long to fix, yet it’s advertised as the future of space travel. Also, let’s talk about Elon himself. You really want our future to belong to a Reddit bro? He’s openly anti-Semitic, has 10 children among three women, half of whom won’t speak to him. But no, he can run a colony. He can’t even keep his own family together, and somehow you expect to run a city on another planet. People just listen to anyone with money.

Nice sensationalist tabloid summary.  Anyway...

36 minutes ago, Kerbalsaurus said:

But like… Starship has only had 3 orbital flights in the span of 11 months. Not as rapid as the F9, which has flown 27 flights since the new year.

You are reading incorrectly what was written, twice.  At the same point in F9 development, Starship has faster turnaround.  Not comparing to F9 turnaround now.  I truly hope this helps

Edited by darthgently
Link to comment
Share on other sites

3 hours ago, Kerbalsaurus said:

But like… Starship has only had 3 orbital flights in the span of 11 months. Not as rapid as the F9, which has flown 27 flights since the new year.

Falcon 9 when it was NEW.

They flew twice in 2010 (first flight was 2010).

The next flight was mid 2012, and they flew twice in 2012, and twice in 2013.

2013:

 

 

 

 

Edited by tater
Link to comment
Share on other sites

That's got a lot to unpack. If my take is worth anything, I offer it.

 

47 minutes ago, Kerbalsaurus said:

But Starship isn’t exactly reliable. It fails almost every flight

My dispute is that Starship was never really expected to be reliable three flights in. Arguably, in terms of flight count, it is doing better than expected.

Starship is arguably aiming a lot higher than any other rocket in history. They want a cheap, reusable, rapid turnaround platform that can refuel in space and send people to the moon and Mars. That was widely considered a pipe dream upon announcement in 2016 and still is by many. Nobody thought they would get it right, let alone the first try. SpaceX knew this. You can't make something that revolutionary and expect it to be perfect the first time around. The space shuttle, a much less ambitious system, put a LOT of effort into doing things right the first time, and just barely succeeded (the mission anomalies section on the STS-1 Wikipedia is so long that I don't want to go through it here). The Littoral Combat Ship program is another example of this, it tried to incorporate a laundry list of revolutionary technologies at the same time, and now a lot of those brand new ships are being decommissioned because they don't work.

SpaceX understands that you can't do something that revolutionary and have it go right the first time, so instead of putting all of their eggs in one basket, they built not just iteration, but rapid iteration, into the design process. The Space Shuttle was crew rated from the start, and each vehicle was so expensive that any design changes you make can't be done by scrapping and rebuilding, they have to be built into the vehicle. Thus, if you found out something was wrong with it, your options for fixing the problem were rather limited. The LCS program is working on an improved second batch of ships, but they are coming along fairly slowly.

This focus on iteration also allowed them to be a bit riskier with their design decisions early on. Ideas that might work were actually tested, as the worst that could happen was waiting for the next vehicle to be done. While exactly how risky is the best amount of risky is a topic of contention, they have learned a lot about what is needed vs what isn't needed, and disproved some assumptions nobody has really challenged since the beginning of the space race.

The end result of this is that immediate success is not the expected outcome. Three failures is eyebrow raising but not necessarily the end for any other conventionally developed rocket. Falcon 1 failed 3 times before succeeding and look where SpaceX is now. The Firefly Alpha failed twice before its first success (then promptly failed again) (admittedly 2 partial failures which placed the satellites too low to last more than  a few weeks), and nobody is saying Firefly is dying.

As for how many failures it takes to doom a rocket, Astra's Rocket 3 failed four times (six if you count suborbital tests) before getting to orbit. It then proceeded to have 2 more failures and 1 more success, failing on 3/5 of its customer serving launches before going bankrupt. That is what is required to doom a rocket if you don't have strong financial backing.

3 failures before succeeding is not that unexpected for any normal rocket, much less the most ambitious rocket in history which was intentionally developed in a way where more failures than average were expected.

 

1 hour ago, Kerbalsaurus said:

and frankly I don’t even know why they’re shooting for orbit, considering it could barely go to sub orbit.

My dispute is with the idea that Starship can only barely go to sub orbit, and with the idea that there is a meaningful engineering difference between a vehicle that can reach the type of sub-orbit they are doing and an actual orbit.

Had IFT-3 been a Starlink deployment flight it likely would have succeeded in deploying Starlinks. It made it to the planned trajectory (the only difference between it and orbit being a few seconds at most of engine burning time), and opened the door. It was later unable to control itself and was destroyed on re-entry, which is something that nearly every other rocket does by default. IFT-3 showed that Starship is capable of doing pretty much everything a baseline expendable rocket can do, the only issues were in space restart (which not all expendable rockets do) and long term attitude control (which not all expendable rockets do). The only reason the mission profile wasn't 100% of the way to orbit rather than 99% was likely because they were concerned about their ability to restart the engine and wanted to prove they could do that before needing to.

This profile allows them to collect re-entry data even if the engine fails (or would have if attitude control didn't fail), and avoids another Long March 5 incident - a few years back on the early Long March 5 launches, the core stage makes it all the way to orbit on purpose and passively de-orbited. As the landing site was unknown, it caused a bit of a panic as it was big enough for parts to reach the ground. A larger and more robust vehicle already designed to survive re-entry would be far worse.

I do know why they are shooting for orbit (or near orbit at least), they have proven they can do the prerequisite steps and are ready to test out re-entry and in space operations.

 

1 hour ago, Kerbalsaurus said:

And Starship is taking so long to fix,

My dispute is with what the expected fix duration is. May I ask what you are comparing this to? Seems pretty fast to me.

There are two parts to this, turnaround time between launches and development time.

Starship had 212 days between flights 1 and 2, and 117 days between flights 2 and 3. That includes the time needed to fix everything both physically and regulatorily.

For Saturn V those numbers were 147 and 261 days. For the N-1 it was 132 and 723 days. SLS has yet to have a second flight after 493 days. While this sample size is too small to determine much, the numbers are in the same order of magnitude.

Comparing to other things SpaceX has done, Starship is very fast. Falcon 1 took a year between 1-2 and almost a year and a half between 2 and 3. Falcon 9 was 7 months from 1 to 2 and a year and a half between 2 and 3. Falcon Heavy was 14 months between 1 and 2, and about 2.5 months between 2 and 3.

 

On the development time side of things, Starship on average is taking about twice as long as expected using the targeted timeline from the Q3 2016 announcement. The details are quite cluttered so are in a spoiler.

Spoiler

Doing the math quarterly because days are a pain.

Ship flight testing was to start in Q3 2018, and ended up starting in Q3 2019 (taking the 150m Starhopper flight as a conservatively late start time), taking 1.5x as long as expected.

Orbital testing was to start in Q1 2021 and ended up starting in Q2 2024, taking 1.75x as long as expected.

Booster flight testing was to start in Q2 2019 and never ended up happening. Taking a conservatively late number to help your case, the first attempted all engines static fire of a booster took place in Q1 2023, taking 2.35x as long as expected.

Ship flight testing was to end in Q4 2020 and ended up ending in Q2 2021, taking 1.1x as long as expected.

Booster testing was to end in Q1 2021 and hasn't ended yet as it is implied to extend into the orbital test campaign. I would say booster testing ends once landing is achieved. It has so far taken at least 2x as long as expected.

Orbital testing was to end in Q3 2022 and hasn't ended yet. It has taken at least 1.25x as long as expected.

Using values of 3x and 2x for booster and orbital testing end dates, the average of all milestones excluding Mars comes out to 1.95, or roughly 2x schedule slip.

A 2x schedule slip isn't unusual. 

SLS was funded in 2011 and targeted a 2016 debut, taking 2.2x as long as expected.

JWST design started in 1999 targeting a 2007 launch. While there were a lot of redesigns, taken at face value this is a delay factor of nearly 3x.

The state of Falcon Heavy was nebulous for a long time, but was mentioned in 2008, and by 2011 they were targeting a first flight in 2013. This is a delay factor of 2x to its first flight in 2018.

I'm tired of doing math but I remember a time where New Glenn, Ariane 6, Vulcan, and H-3 were expected to be online in 2020. Just now in 2024 we finally have 2 of those.

 

Maybe the reusability program is taking longer than expected?  Falcon 9 is probably the closest thing we have.

Falcon 9 did 13 low altitude tests (grasshopper and F9R dev) plus 9 various full up landing tests before successfully landing a booster.

So far, Starship has done 5 low altitude tests, 5 medium altitude tests, and 1 full up ocean landing test, for a more ambitious vehicle. The booster has done 1 fill up landing test and no hop tests. It is difficult to gauge how far along we are in comparison, but in terms of test count, they appear to be getting closer to flight readiness with fewer flights. Assuming it takes 12 flights to get to a successful booster catch and land landing of the ship, that's 23 Starship test flights compared to the much simpler Falcon's 22.

 

2 hours ago, Kerbalsaurus said:

yet it’s advertised as the future of space travel.

My dispute: Every space agency/company advertises their rockets as the future of space travel.

The shuttle was billed as a low cost rapidly reusable space tug, the SLS was supposed to be NASA's low cost flagship rocket for the next several decades, Rocket Lab and Astra were talking a hundred flights per year... On the positive end, Cygnus and Dragon were billed as low cost shuttle replacements and worked. The shuttle was indeed the future of space travel, covering several decades for better or worse. And Falcon 9 nearly reached 100 flights last year.

But more relevantly, any moderate cost low refurbishment fully reusable vehicle is going to be the future of space travel. I am not convinced that in 50 years we are going to throw away our rockets after one flight like we are still doing, even with Falcon's second stage. Starship and Terran R are the only two rockets seriously aiming for that future, and of the two Starship is by far the closest one. If any rocket can claim to be the future, I think we have a winner.

 

2 hours ago, Kerbalsaurus said:

Also, let’s talk about Elon himself.

I won't go too much into detail because this is probably going to get moderated if we go much further, but I do frequently have major concerns about what could happen if Mars is run the way SpaceX is run. Even just how SpaceX is run. A friend of mine has a final interview there tomorrow and I am both excided for and concerned for her. I can see people accepting a high stress high reward environment like that for a couple years without a family (I will likely apply there soon, admittedly I am a bit behind on the graduation job hunt), and obviously the first Mars missions will be high stress, high risk, high reward positions crewed by professionals. But long term, Mars has to be a place for everyone. Musk isn't the only one that would be involved with the colony, though, and will likely be dead before a Mars colony gets going.

For what it is worth, while it may not be fully convincing, he has condemned the tweet you are referencing as the most foolish thing he has ever said.

 

Link to comment
Share on other sites

Just now, Vanamonde said:

This one's for Artemis. 

Direct message from someone at NASA texted to me while we were both watching IFT-3:

Quote

A rather well known GC said while we were looking at that video from orbit, “we are watching the end of SLS”

In terms of the Starship discussion WRT alternate Artemis architectures—they're not gonna happen unless NASA as a motivated customer actually asks for it.

ObArtemis:

GJS5JcTWUAAkJlu?format=jpg&name=4096x409

 

Link to comment
Share on other sites

26 minutes ago, GuessingEveryDay said:

 

In terms of looks, Intuitive Machines wins for me, it just looks so dang cool. It's more likely that Lunar Outpost could win for cargo.

That rover really loves!!! ( in the good sense)

also I love how they called it "moon racer".

the 2nd one looks like it is more a pickup for heavy duty stuff

the 3rd is a light quadricycle

Link to comment
Share on other sites

Intuitive Machines' Moon Racer: near-future Aristarchus 500 rally entrant

Lockheed-Martin's Lunar Dawn: TRUCK. *grunts in American*

Astrolab/Axiom: About to drop some Lunar beats

Link to comment
Share on other sites

22 minutes ago, AckSed said:

Intuitive Machines' Moon Racer: near-future Aristarchus 500 rally entrant

Lockheed-Martin's Lunar Dawn: TRUCK. *grunts in American*

Astrolab/Axiom: About to drop some Lunar beats

Tesla really needs to lunar-harden CyberTruck

 

Edited by darthgently
Link to comment
Share on other sites

5 hours ago, DAL59 said:

The REAL best Artemis architecture

BDB and ACK threads need to see this.

Also, this would theoretically be more effective and cost less than most of the real proposals.

Link to comment
Share on other sites

I can’t post the link because the article contains a mix of space and off topic political news, but CBS News reports Japan and the US have announced that two Japanese astronauts will fly on Artemis missions, and, one of them “will become the first non-American to land on the Moon.”

As a Japanese person I’m extremely excited about this. On the other hand, apologies to any European members of the forum who may be disappointed.

A single caveat: the Artemis cadence is garbage. If Ars Technica’s prediction of an Artemis III mission in 2028 is correct, and it takes a couple years to fly Artemis IV, it’s somewhat possible China will land people on the Moon before then. I suppose it’s possible a Japanese astronaut would fly on Artemis III, but who knows.

Link to comment
Share on other sites

24 minutes ago, SunlitZelkova said:

I can’t post the link because the article contains a mix of space and off topic political news, but CBS News reports Japan and the US have announced that two Japanese astronauts will fly on Artemis missions, and, one of them “will become the first non-American to land on the Moon.”

I saw something that suggested a Moon Cruiser pressurized rover, but was looking for something with more info... Toyota was involved, right (Land Cruiser—>Moon Cruiser :D ).

That might be a more substantial contribution than the Orion SM... which uses Shuttle engines, so it's just some tanks, etc, and a few hundred million. I have to imagine a pressurized rover is billions (certainly would be if NASA did it, they've spent that kind of $ already on surface EVA spacesuits that don't exist yet).

Link to comment
Share on other sites

https://global.toyota/en/mobility/technology/lunarcruiser/

Quote

The 6 meters long 5.2 meters wide, and 3.8 meters high pressurized rover’s body will entail four core technologies: regenerative fuel cells, automated off-road driving, off-road driving performance and user experience. Despite the moon’s two weeks days and nights2, the regenerative fuel cell (RFC) will allow the vehicle to establish an energy cycle for sustainable and long-term moon exploration through water electrolysis and fuel cells. Moreover, to increase off-road driving performance, the rover will be equipped with motors and independent steering mechanisms in each of its wheels

Volume is maybe 80-90 m3 (larger than Shuttle's). It is supposed to be able to support crew for 30-45 DAYS.

A few observations.

1. This means that the amount of life support gear to do this is fairly compact (as a constraint on HLS... is not much of a constraint). Crew could easily be 4 in that volume.

2. The uncrewed autonomy means that assuming it's built for resupply, any lander can land, then rover drives up to it, the new crew swaps out whatever... and off they go, exploring. Drive back, enter HLS, rover drives behind that hill over there for safety... it solves some logistics problems about fixed infrastructure and ejecta damage from landers.

 

Link to comment
Share on other sites

1 hour ago, tater said:

Drive back, enter HLS, rover drives behind that hill over there for safety... it solves some logistics problems about fixed infrastructure and ejecta damage from landers.

It was my understanding that the thrusters on the upper half of the vehicle were meant to be used for final descent and thus avoid problems with spewing regolith everywhere.

In any case, they'd still probably want to park behind a hill just in case of the lander exploding or something.

Link to comment
Share on other sites

21 minutes ago, SunlitZelkova said:

It was my understanding that the thrusters on the upper half of the vehicle were meant to be used for final descent and thus avoid problems with spewing regolith everywhere.

In any case, they'd still probably want to park behind a hill just in case of the lander exploding or something.

Yes. But the second contract in HLS is the Blue Origin lander, so even if the landing engines stay high on LSS there's a concern I think.

I am using HLS generically for both.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...