Evil_Merlin

Ars Legatus Legionis
23,745
Subscriptor
Screenshot 2025-08-26 133046.png
 

Attachments

  • Screenshot 2025-08-26 133333.png
    Screenshot 2025-08-26 133333.png
    333.9 KB · Views: 32

xoa

Ars Legatus Legionis
12,364
Subscriptor++
Yeah, that was beautiful start to finish. I think everyone having been able to watch from Starhopper onwards adds something too, the contrast from the first stumbling steps until now is a different feel from just seeing a final polished product go. Most importantly looks like they hit every single data gathering goal, which should give them a lot to work with for V3 and really getting Starship to start to pay for itself. Not that it's particularly surprising but great to see the work pay off.
 

Ecmaster76

Ars Legatus Legionis
16,977
Subscriptor
Nice to see a return to confident attitude control throughout

The pez dispenser in action makes me think someone at SpaceX can or has designed a shipping pallet cannon

It was a funny coincidence that the camera switched to the engine bay view and the announcer said "what really helps us punch through" right before the skirt popped a hole. Thats a spot where they have previously done missing tile tests which makes me wonder if they were expecting it then.
 

Chito

Ars Praefectus
4,076
Subscriptor++
Maybe I missed something along the way, but does Super Heavy not do/need a re-entry burn like Falcon 9's first stage does? There was just a boostback burn and a landing burn, seemingly with the booster pointed engines-forward most of the way down.
Their trajectory was suborbital this flight in case of big booms and failures. So no burn needed.
 
  • Like
Reactions: zeotherm

BigP

Ars Praetorian
459
Subscriptor++
Their trajectory was suborbital this flight in case of big booms and failures. So no burn needed.
I don't think you have that quite right. The booster (Super Heavy) does not do a re-entry burn by design, unlike F9. I'd have to compare launch videos or find a reference to be sure but I think SH stages lower and slower than F9 and with its larger diameter should have a larger stand-off distance from the shockwave/boundary as it comes down. Watching the re-entry heating of that area is quite spectacular though!
 
  • Like
Reactions: MilleniX

Chito

Ars Praefectus
4,076
Subscriptor++
I don't think you have that quite right. The booster (Super Heavy) does not do a re-entry burn by design, unlike F9. I'd have to compare launch videos or find a reference to be sure but I think SH stages lower and slower than F9 and with its larger diameter should have a larger stand-off distance from the shockwave/boundary as it comes down. Watching the re-entry heating of that area is quite spectacular though!
You are right, I was thinking of the ship not the booster. Shouldn't sleep and post ;).
 
I don't think you have that quite right. The booster (Super Heavy) does not do a re-entry burn by design, unlike F9. I'd have to compare launch videos or find a reference to be sure but I think SH stages lower and slower than F9 and with its larger diameter should have a larger stand-off distance from the shockwave/boundary as it comes down. Watching the re-entry heating of that area is quite spectacular though!
There are at least a few factors to consider:
  • Raptor/Super Heavy had a specific design goal to withstand higher temperatures than Merlin/F9
  • A current F9 booster is expected to be re-used 10-20 times, a Super Heavy just once or twice (at least during development). A bit of erosion which is irrelevant for 1-2 launches could be catastrophic for 10+
  • Super Heavy/Starship are still deliberately being pushed to find (and iterated to expand) the edges of the design envelope. Raptor 3 is much more heat tolerant than Raptor 2, which is more heat tolerant than Raptor 1.
 

BigP

Ars Praetorian
459
Subscriptor++
Oof. Yeah, I completely forgot to mention the materials difference (steel for SH, aluminum (inconel alloy?) for F9). Steel is significantly more heat tolerant. I'm looking forward to seeing how the Raptor 3s handle being more exposed as well.

I don't know that I'd go so far as to say that current SH boosters are deliberately less robust in the designed reuse cycle life sense right now though, but that's certainly possible. I've seen nothing on way or the other. (But to be fair, my enthusiasm for looking closely has been lower than in years past for some damn reason. . .)
 

continuum

Ars Legatus Legionis
97,602
Moderator

MilleniX

Ars Tribunus Angusticlavius
7,786
Subscriptor++
Their trajectory was suborbital this flight in case of big booms and failures. So no burn needed.
It was sub-orbital only in the sense that its perigee was well inside the atmosphere. It had as much kinetic energy as any other LEO launch - about 7 km/s velocity at shutdown.

But regardless, I was talking about the booster, not the ship, as noted.
 
  • Like
Reactions: Xanrael

QtDevSvr

Ars Legatus Legionis
12,196
Subscriptor++
Starlink sims going out the door now. Very smooth.
Yeah I thought the sim deploy on the last mission was a good example of their "minimize the hardware" approach. Deploy was rickety to the point where it looked like things could jam or otherwise malfunction. They beefed things up this time, with good data on just how much more engineering might be needed, and the deployment was very smooth.
 

xoa

Ars Legatus Legionis
12,364
Subscriptor++
Last night, B18 blew out the side of the aft end, possibly another COPV bursting. Initial pressure testing, not involving propellants, so no fire.
I'm seeing some places saying it was a LOX tank structural failure, hopefully they do another post-mortem.
Don't think this one is going to buff out.
Nope, though I'm having a little trouble telling whether this is that significant or not. Looks like stacking for B18 took about 170 days, which obviously would be quite a delay. But stacking B17 only took around 60 days, B16 took 71 days, and B15 took 52 days, so it seems like the extended time there was more likely down to changes and experimentation they were doing vs Boosters taking that long normally. And B19 was apparently spotted beginning production back in August (though depending on the failure mode B19 might also have the same thing).

So I guess it'll matter what it was? If it was a COPV failure, that's simultaneously bad & good, bad because it's another one in something that should be thoroughly proven at this point and either likely cause (manufacturing fuckup or handling fuckup) points to sloppiness to get under control. But good in that it's a pretty independent component, there aren't any design implications for screwing up with COPVs they just need to get that under control one way or another. In this scenario presumably B19 is only delayed by efforts to fix that, because the COPVs themselves are just drop in components and wouldn't be changing in dimensions just, like, not dropping them or miswrapping them or whatever.

If it's a structural failure in something they've designed and altered for Booster 3 though, then the good side might be in that it's discovering a new thing which of course is the point of testing. I guess it'd make me wonder if perhaps they've started to embark on envelope hunting for mass savings, if they began pretty fat with extra safety margin, as they head for production maybe they'll be looking at where they can cut and miscalculated how far? But that could also mean some more significant design reworks are required, and if that has upstream/downstream effects it might mean B19 is also useless.

Anyway, bummer but better this than a big boom while fueled.
 

Ecmaster76

Ars Legatus Legionis
16,977
Subscriptor

MilleniX

Ars Tribunus Angusticlavius
7,786
Subscriptor++
On the SpaceX reddit, somebody claimed it was a procedural SNAFU. A vacuum was pulled somewhere it really shouldn't have been. That's just an anonymous post, take with salt.
Wouldn't be the first time SpaceX has been caught out by a testing procedure creating catastrophic transient states. You'd think they'd have learned to do at least some minimal simulation before risking real hardware, but maybe that truly is just not how they roll. Live by the "hardware rich" testing, die by it, I guess.