totm nov 2023 SpaceX Discussion Thread

[tater]

 

"At SpaceX we specialize... in converting things from... 'impossible' to 'late.'"

Right now at landing SH is ~250t, they think they can get it under 200t.

[tater]

Starlink 2 sat is 7m long.

~1250 kg

Almost an order of mag more capable than starlink 1.

[Ricktoberfest]

22 hours ago, tater said:

I want a pad cam like that one they did for Starship... looking right up the engine bells as it lands.

Hmm, for some reason when I search “spacex upskirt video” I don’t find any landing pad videos.

[RCgothic]

 

 

[sevenperforce]

1 hour ago, Ricktoberfest said:

Hmm, for some reason when I search “spacex upskirt video” I don’t find any landing pad videos.

I just keep seeing landing strips

[RCgothic]

4 hours ago, tater said:

Starlink 2 sat is 7m long.

~1250 kg

Almost an order of mag more capable than starlink 1.

I think this was the juiciest factoid to come out of that interview.

[Beccab]

Road is closed & tank farm venting, so S24 may be prepared for cryo already

Edited May 27, 2022 by Beccab

[Beccab]

 

[tater]

Proposed space telescope constellation (8m objective!) launched by Starship:

https://nautilus-array.space/

Cool to see ideas using the available volume.

Edited May 27, 2022 by tater

[Beccab]

 

[StrandedonEarth]

Now what did they break?

 

[RCgothic]

Yeah, it wasn't great.

 

 

[Pthigrivi]

Can we all agree this is either the smartest thing ever or the stupidest?

Edited May 29, 2022 by Pthigrivi

[SunlitZelkova]

7 hours ago, Pthigrivi said:

Can we all agree this is either the smartest thing ever or the stupidest. 

We don’t know the details. I would say “normal” as they are still testing and as much as Musk likes to talk about simplification a super heavy lift launch vehicle is a complex thing.

Even if we get details saying it was a dumb human error or something I would inclined to say it’s alright, because SpaceX’s development philosophy encourages such incidents. Unlike Boeing which claims their method will lead to guaranteed success and then suffers a variety of issues.

[Pthigrivi]

56 minutes ago, SunlitZelkova said:

We don’t know the details. I would say “normal” as they are still testing and as much as Musk likes to talk about simplification a super heavy lift launch vehicle is a complex thing.

Even if we get details saying it was a dumb human error or something I would inclined to say it’s alright, because SpaceX’s development philosophy encourages such incidents. Unlike Boeing which claims their method will lead to guaranteed success and then suffers a variety of issues.

Oh no sorry I mean trying to catch this thing with robot arms. Absolutely nuts haha

[GuessingEveryDay]

13 minutes ago, Pthigrivi said:

Oh no sorry I mean trying to catch this thing with robot arms. Absolutely nuts haha

Eh, Super Heavy has the potential to hover, so it really depends on how quickly the arms can move laterally. Plus the winds, although I think the engines can manage that. We'll probably won't see them trying to land on the Chopsticks with a Super Heavy until B10 or after that.

[Ultimate Steve]

I initially thought it was stupid, but the more and more I think about it, the more I'm convinced it is a good idea. If a rocket is to be fully and rapidly reusable, traditionally, it would either:

1. Land nearby, be craned back onto the launch pad
2. Land directly on the launch pad
3. Land somewhere non specific, be refueled, and launched again from the landing location
4. The SSTO spaceplane approach

2 would be ideal, and would not require much, if any landing gear, but requires insane accuracy. 3 would remove the accuracy problem, but would require very heavy landing gear to support the weight of the fully refueled vehicle for liftoff, and you would still need stacking equipment for the upper stage if it isn't an SSTO, and even then, the payload needs to be loaded somehow. 1 requires time and stage transportation infrastructure as well as (lighter) landing gear.

The mechazilla approach, if it works, relaxes the accuracy requirements (depending on how far and in what ways the chopsticks can move), minimizes transport times (needs to be craned back onto the pad but the thing lands in the crane and is a few dozen meters above the pad at most), and allows the deletion of the entire landing gear system. Some structural reinforcement is needed. The current system uses the same lift points as would normally be used, so if done correctly, not much extra reinforcement is needed.

It does add a lot of new failure points, and there are no do-overs for the catch (but tbf there aren't any do overs for any of the other landing methods), but the robot is not subject to the strict mass requirements of a rocket, and can be designed with a ton of redundancy. It also removes the failure points of a reusable rapidly retractable landing gear system over approaches 1 and 3. 

This is a comparatively minor bonus, but the mechazilla approach removes (if done well) the possibility of engine damage to the launch complex, and FOD damage to the booster due to how high the booster is off the ground when it is caught.

Now, landing the *ship* in the chopsticks, I'm still not convinced on, but I might be slowly warming up to that, too. Getting the landing location consistent given how many different engine out scenarios there are for the flip maneuver... That's a tall order. If the entire belly wasn't coated in ceramic tiles, I would think that I would prefer the "land in a giant net" approach for the ship, but for now, I still prever normal vanilla "land the ship on a landing pad" especially because the payload has to be loaded somehow. Now, for rapid refueling of ships in orbit, landing on the chopsticks is indeed attractive as the fuel can just be pumped like it is normally.

 

[mikegarrison]

Yeah, it's pretty obvious that if they really want to have airline-style rapid turnaround, the best way is to land the rocket right back on the launch site. Airplanes land and take off on the same runways.

However, it's risky AF. If anything goes wrong at the last moment, you not only lose your rocket but also damage your landing and launching infrastructure. Their experience with Falcon 9 seems to show that they can almost always land their rocket right in the middle of the landing circle, which probably means they have a decent chance of having enough control to do this catch thing. But the consequences of a landing failure are a lot more severe because the collateral damage can be quite a bit higher.

[magnemoe]

22 hours ago, Ultimate Steve said:

I initially thought it was stupid, but the more and more I think about it, the more I'm convinced it is a good idea. If a rocket is to be fully and rapidly reusable, traditionally, it would either:

1. Land nearby, be craned back onto the launch pad
2. Land directly on the launch pad
3. Land somewhere non specific, be refueled, and launched again from the landing location
4. The SSTO spaceplane approachf engine damage to the launch complex, and FOD damage to the booster due to how high the booster is off the ground when it is caught.

1 is the safe method, pretty much an falcon 9 RTLS.
2 is harder as you need to hit an smaller target and missing will damage the launch pad, the landing system will also be different from the launch one. 
3 Does not solve the problem, now it could be relevant for starship because overflight. 
4 its the old idea of reuse, problem is that its expensive to develop and adds lots of dry mass over 1, benefit of powered landing is that yes you need more fuel but this fuel is also an backup if you get problems, you can then ditch the first stage to save the second. 

Now I would not gone directly for the chopsticks. I would add some simple legs and train on landing very accurately, perhaps jump with superheavy alone to try out this. 

[sevenperforce]

2 hours ago, magnemoe said:

Now I would not gone directly for the chopsticks. I would add some simple legs and train on landing very accurately, perhaps jump with superheavy alone to try out this. 

They won’t try to catch Superheavy the first time; they’re dropping the first one in the Gulf so they’ll use that to assess landing accuracy. 

[Beccab]

S24 is also having the broken pipe replaced right now

[Ricktoberfest]

On 5/27/2022 at 11:56 AM, tater said:

Proposed space telescope constellation (8m objective!) launched by Starship:

https://nautilus-array.space/

Cool to see ideas using the available volume.

I keep seeing this layout, but I wonder if it would be better if the load were attached to the inside of the clamshell door?  Then they just have to slide off guide rail instead of at a weird angle.   Bad part would be the door would have to be able to support the load during high G launch

[CBase]

10 hours ago, Ricktoberfest said:

I keep seeing this layout, but I wonder if it would be better if the load were attached to the inside of the clamshell door?  Then they just have to slide off guide rail instead of at a weird angle.   Bad part would be the door would have to be able to support the load during high G launch

Maybe rather vice versa: attach the door to the payload rack like kitchen appliances. With a 30° - 45° opening it should be open enough to let the payload slide out. Anyway stress on the door is mainly on ascend when it can be manually closed after loading. On return from orbit even some small gaps on backside shouldn't matter - the pez dispenser door could propably even stay open -, which does simplify whatever construction they will head for.

[magnemoe]

2 hours ago, CBase said:

Maybe rather vice versa: attach the door to the payload rack like kitchen appliances. With a 30° - 45° opening it should be open enough to let the payload slide out. Anyway stress on the door is mainly on ascend when it can be manually closed after loading. On return from orbit even some small gaps on backside shouldn't matter - the pez dispenser door could propably even stay open -, which does simplify whatever construction they will head for.

Main problem with the wale mouth door is loading payload as I see it, this get even more problematic if you can not open the door fully.  
In orbit its main problem is that you can only deploy from the top. 

The pez dispenser on the other hand is smart.  The only issue I see with it is the danger of the stack of large thin satellites jams in the rack. 

Standard way to lock large hatches is to use rows of clamps usually controlled by an motorized rods who locks the door so well that large cargo hatches on planes airtight.  This will not be needed on cargo starship but adding some structural strength would be nice. 

[RCgothic]

Another 2 week delay: