Continuing discussion of SpaceX launches [edited title]

My wife and I honeymooned in Florida, and we specifically timed our honeymoon so we’d be near Titusville foe a shuttle launch. It was STS-31 - the mission that launched the Hubble.

We got a great spot to watch the launch from just across the water, and it was very exciting as the countdown progressed.

That is, until the launch was scrubbed at T-4 minutes. It didn’t launch for 14 days, after we had gone home. Sigh. I still haven’t seen a rocket launch in person.

I’ve seen one Space Shuttle and one Falcon launch while flying up or down the FL peninsula. Both were impressive even from ~6 miles up and ~40 miles away. Their short time to climb to our altitude is astounding. And we’re near the top of our vertical range and they’re just barely out of their starting gate when they go by.

While I was stationed at Patrick AFB on the coast some 15 miles south of Canaveral I watched an evening launch of something from the beach at the base. I can’t recall what it was, but this was mid-late 1982, so almost certainly a flavor of Delta or Atlas putting something into orbit for DoD. Or possibly for NASA, but DoD was the way to bet.

I grew up in Florida, partly; we moved there when I was 10 or 11.

I saw the first shuttle launch from the Burger King parking lot that was across the water from the launch pad. I’ve been on and in the MCTF (I sat in each of the four drivers’ seats!). Over the years I saw dozens and dozens of launches, some from “just over there” close and some from 50 miles away.

A rocket launch in person is truly awesome.

SpaceX lands their 100th booster:

Their fleet leader has 11 launches/landings to its name, and they had a 30/31 landing success rate this year.

I think we’re finally past the point of goalpost moving, though. The last batch of arguments were basically “it’s only economic if you can reuse 10 times” (not actually true) and “it’ll never be reliable enough”. Well, obviously those issues aren’t a problem for SpaceX, and no one is bothering to shift the goalposts further. France at least is being very explicit:

“For the first time Europe … will have access to a reusable launcher,” Le Maire said, according to Reuters. “In other words, we will have our SpaceX, we will have our Falcon 9. We will make up for a bad strategic choice made 10 years ago.”

Good luck to them achieving that with a bureaucratic organization like ArianeGroup, but good on them for recognizing their mistakes.

Nothing particularly special about this mission, but they had some great ground photography this time:

I can’t recall ever seeing stage separation from the outside with that level of clarity. Normally you just get a few overexposed pixels. Unfortunately, they cut away the outside shot at the moment of fairing deploy, but you do get to see the fairing halves spinning off.

Musk is doing a Starship update in about a half-hour:

And by a half-hour, I mean 30 minutes plus however long it takes him to show up. So, let’s say one hour.

If you watched, were there any highlights or surprises? I started to watch and he was rambling all over and I lost interest.

Not much new factual information, but some nice videos and photos. Raptor 2 looks to be a very nice improvement over the original, and is apparently a full redesign (they’re still working on not melting the main chamber).

They think that both the FAA environmental review and the hardware will probably be ready in the next couple of months. Musk doesn’t think there will be any huge holdup on the environmental review, since he considers everything to be fairly non-impactful at an objective level, so presumably it’s just a matter of convincing the FAA. However, on the off chance that there is a big holdup, they will move the launch operations to Florida, where they already have approval. That will add a delay, but they’re planning on starting operations there anyway, so it won’t be a huge deal.

The launch tower in operation is very impressive. The “chopsticks” can move and place the Starship stage, while the arm below it (containing the QD [Quick Disconnect {for the propellants}]) system can clamp onto the booster, stabilizing it. It’s pretty incredible to see this 400-foot articulated robot thing moving around like that.

Overall, Musk seems pretty optimistic. He didn’t mention any serious obstacles to orbital flight. He doesn’t expect the system to necessarily work on the first try, though he thinks they’ll nail the landings and such faster than Falcon 9. They’ve built a production system with huge capacity that will mean they can easily deal with a few failures.

They expect to reuse the booster in under an hour. 6 minutes of flight, 30 minutes of prop load, and some stacking time. The orbiter takes longer due to orbital mechanics–depending on the orbit, it may take 1-3 full orbits to line up again for landing. So it takes at least 8 hours to cycle. Hence, they’re shooting for a Starship production rate about 10x that of the Super Heavy.

Any particular things you were interested in?

Thanks large for the rundown. I was mostly interested in the launch schedule for a full-stack Starship with heavy booster launch. Where would you say we’re at with that?

Anywhere from two months to the end of the year is what I gathered. Yeah, that’s a pretty big range, so I’m presuming the two months is something like a 50/50 shot, while the end of the year is more like 90%. The hardware really looks to be coming together, but it’s definitely not done, so lots of things can happen. Not to mention the environmental review.

So probably this fall. That is damned cool.

Yeah, fall seems like a pretty good bet. It’s gonna be exciting!

BTW, I’m not sure how closely you’ve been following Starship development in general, so when I say there’s not too much new, that’s relative to the knowledge that an army of onlookers have already seen down in Texas with several 24/7 cams, drones, and so on. So it may be that you (or any other readers!) have more basic questions that I’m leaving unaddressed. If so, ask away!

Raptor 2 on the left, Raptor 1 on the right:

Raptor 2 has 230 t thrust as compared to 185 t, despite being vastly simpler. Though they’re still working on keeping the combustion chamber from melting…

Musk mentioned that he hates flanges. They’re heavy and weak. And while he didn’t explicitly say it, I think another aspect of his design philosophy is coming into play: aside from manufacturability concerns, you only need a flange if you expect to take the thing apart at some point. But if it’s reliable enough, you don’t need to take it apart. So delete the flange. I still see some flanges on Raptor 2, so I expect a Raptor 3 when they work out a few more kinks.

As an aside, Musk has expressed this idea in terms of component testing. Component testing makes sense when all your components are unreliable. But eventually, you’ll have worked out the bugs. You should switch to just testing the final product at that point and stop wasting time with the components. However, testing sometimes gets embedded in the manufacturing system and sticks around as waste until someone explicitly rips it out.

And really, it’s the same idea behind pooled testing for (say) COVID. Use individual testing when the prevalence is high. But when it’s low, it’s a waste to test everyone. So pool the tests and only do something extra on the rare positives.

Quite a good podcast episode from the Economist about Starship here:

And it also looks like SpaceX are beginning construction of a Starship launch tower at Kennedy Space Centre

I do feel the FAA might limit SpaceX’s permission to use Boca Chica for Starship. Elon started talking about using it for R&D and I think all signs point to them anticipating an unfavourable decision from the ongoing environmental assessment.

A good article on the SpaceX reusability efforts:

Nothing too surprising in there for SpaceX enthusiasts, but it’s nice to hear about all the little details that have led to their increased flight cadence.

Gerstenmaier worked for NASA for 42 years, and then SpaceX. He found that it was virtually impossible to remove requirements in the Shuttle program, but at SpaceX it’s just part of their iterative process.

Of course, the Falcon 9 has the distinct advantage in that most of its missions are unmanned, and so they can afford to take more risks. The information gained from those missions can be fed back into all of their other missions, increasing the safety of manned missions as well. The Starlink missions especially can take on significant risks, since there is not even an external customer in that case.

I wonder how much of this information is fed back to industry. In particular, the early fairing deployment, which increases the heating rate of the satellites. Dumping the fairing early increases payload capacity and may even increase the safety margin due to having more propellant in reserve. If the satellite is unaffected, why not?

Payload heating analysis has been a critical part of every space launch mission I have ever worked and is often heavily scrutinized during design reviews. The obvious advantages of deploying the fairing as early as possible—especially if vehicle performance to the desired orbital injection point—have to be weighed against the very restrictive heating limits of many payloads because satellites and spacecraft are generally delicate, and heating during ascent not only contributes to the thermal load it has to cope with upon deployment due to solar exposure but can also exacerbate surface ionization, propensity for coronal discharge, and erosion of exposed insulation and antennas.

This is the kind of thing that drives me a little nuts because space enthusiasts act like trying to deploy the fairing or shroud as early in flight as possible is some kind of new idea that SpaceX came up with when it is something every space launch contractor, mission assurance provider, and review board has looked at since the dawn of spaceflight. It’s not a remotely original idea or something that only SpaceX does.

Stranger

That’s all fair enough, but there’s generally a gap between theory and practice. If no one wants to risk their payload, how would you ever decrease that gap? SpaceX can accept more risks on Starlink since their launches and payloads are both cheap. The payloads are also interchangeable; the loss of one set is no great loss in the grand scheme.

Fairings are typically deployed at the same time for each mission, except for Starlink where they deploy earlier. Is it really the case that every satellite has exactly the same heating restrictions, or did they simply use a common industry value that everyone knows to target?

I suspect it’s more like the latter case, but that means less opportunity for design trades between payload manufacturers and the launch provider. Trades like this become easier when it’s the same company.

It’s not like I’m making this stuff up out of whole cloth. It’s coming from the SpaceX VP of Falcon engineering. Yeah, they have an interest in tooting their own horn, but it’s not coming from fanboys.

Just to be clear, “same time” means with the same level of heating; not the same actual mission time, which might vary (of course you knew this, but I wanted to be clear).

For reference:

The payload fairing will nominally be deployed when free molecular aero-thermal heating is less than 1,135 W/m2 . Other fairing deployment constraints can be accommodated as a standard service, although they may modestly reduce vehicle performance. Please contact SpaceX regarding mission-unique fairing deployment requirements.

It seems they offer late deployment as an option, but not early.

Fairings are not “typically deployed at the same time for each mission”. It is true that for a common type of vehicle that flies a very typical range of trajectories the fairing deployment is at roughly the same altitude because the dynamic pressure is approximately the same and the atmospheric density tends to drop precipitously at altitude but I’ve worked on missions with very early fairing deployments (around 150 kft) and ones that were only deployed much later (above 250 kft), all depending upon the payload environment and interface requirements as defined by the spacecraft design analysis and qualification testing.

Stranger