totm nov 2023 SpaceX Discussion Thread

[kerbiloid]

3 hours ago, Motokid600 said:

Nah. It'd settle to one side. The landings are much, much softer then parachute. Could even make a cradle to land in.

Legs also keep the nozzles far from ground. Without legs they will fry the fuel tank or explode an engine.
Say, the legless Zarya had its landing engines around the capsule equator, at ~2 m height.
A capsule with near-bottom nozzles needs legs.

(Soyuz's landing engines are on the very bottom, but are much smaller and burn out in less than a second.)

***

Btw, can't they create a reusable propulsion unit for the 2nd stage derived from Crew Dragon?
To drop the fuel tank but to return the engines and command unit.
Say, let it be launched overturned, have a Merlin instead of the docking port.
Or use 8 Superdracos instead of the Merlin, fed from the big single-use tank on top. 8x73 = ~0.57 MN instead of ~0.85 MN.

Edited January 2, 2020 by kerbiloid

[Wjolcz]

2 hours ago, kerbiloid said:

Btw, can't they create a reusable propulsion unit for the 2nd stage derived from Crew Dragon?
To drop the fuel tank but to return the engines and command unit.
Say, let it be launched overturned, have a Merlin instead of the docking port.
Or use 8 Superdracos instead of the Merlin, fed from the big single-use tank on top. 8x73 = ~0.57 MN instead of ~0.85 MN.

They probably could do this but I can imagine it turning into a very bad two-stage-to-orbit with almost no payload capability. They had a similar-ish, but less complex idea in one of their first videos for the F9. 

Edit: I just looked at the comments under the video and there's one from 3 days ago talking about Elon's tweet (presumably from 3 days ago). Did I miss something or did he just post this vid on his Twitter?

Edit 2: I'm going to assume it's the latter.

Edited January 2, 2020 by Wjolcz

[tater]

6 hours ago, Wjolcz said:

Edit 2: I'm going to assume it's the latter.

Someone replied to a 7 year old tweet from Elon with the video in it, and Elon replied to that string of tweets.

 

Edited January 2, 2020 by tater

[Terwin]

On 12/30/2019 at 6:17 PM, CatastrophicFailure said:

 

Soyuz needs retros and Starliner needs airbags... turns out that last bump is a doozy. That’s why SpaceX was originally thinking propulsive landings, better precision and a much softer touchdown. 

If you need compressed air from a COPV to spin up the turbines, how many restarts can you get without an over-sized pressure vessel?  

Is this a case where a 5lb tank(when full) can provide dozens of restarts, or the sort of thing where  you need to start sacrificing cargo space/mass  for each additional restart? 

Is it plausible to re-pressurize the 'starter tank' after use? 

 

On 12/31/2019 at 12:21 PM, Ultimate Steve said:

Also, if starship ends up being like 10 milliion per launch or more initially, it would be cheaper to send a small payload to gto or the moon with a falcon than it would be with a Starship due to the number of starship launches it would take due to refueling.

You could always send a fully fueled vessel inside a cargo starship and have that go to the moon/GTO if you only want to sent up something small, that way it only takes a single(cheaper than Falcon 9) starship launch. 

[kerbiloid]

In that video they land the whole 2nd stage, while I mean separate the CrewDragon-derived propulsion part of it and land on ground like the CrewDragon original idea.
And spend the large cylindric tank part.

[Wjolcz]

The problem with the whole "what if starship is too pricy because of all the refuelling" argument is that F9 can't put 100t into orbit in one go. So if we assume that F9H expendable puts 64t in LEO for 150M$ then if SS puts 100t in LEO (1 launch = 10M$) and it takes 14 launches to refuel it THEN you have 100t in LEO that can be sent to either the Moon or Mars for the price of F9H expendable.

However, I don't know how much fuel the SS holds. I'm about to do research now.

Edit: so I assumed dry mass to be 150t and wet to be 1400t. If the tanker can deliver 100t to LEO then it takes about 13 launches to refill one ship in LEO. So, that's F9Hex price tag for a fully-fueled (AND 100% reusable) SS ready to send 100t wherever its ∆V allows it to.

Edited January 2, 2020 by Wjolcz

[tater]

39 minutes ago, Wjolcz said:

Edit: so I assumed dry mass to be 150t and wet to be 1400t. If the tanker can deliver 100t to LEO then it takes about 13 launches to refill one ship in LEO. So, that's F9Hex price tag for a fully-fueled (AND 100% reusable) SS ready to send 100t wherever its ∆V allows it to.

It's better than that.

A SS "tug" (empty SS, disposable fairing, large docking ring on front in the KSP sense) could send hundreds of tons to TLI. Looks like it has ~4500 m/s dv with a 300t payload docked to the front. That only leaves 700m/s to play with for getting into a LEO, but presumably it could do multiple passes aerobraking, even without a fairing (then the few hundred m/s to circularize).

[CatastrophicFailure]

23 hours ago, tater said:

 

... when the booster failing to disintegrate is considered “off-nominal...” 

[tater]

 

[Motokid600]

15 hours ago, kerbiloid said:

Legs also keep the nozzles far from ground. Without legs they will fry the fuel tank or explode an engine.

I think the clearance difference is negligible. The legs depicted in the early animation with D2 landing were like five inches long. More like pegs then legs.

[Wjolcz]

Do you guys think they will be able to shear off 80t of materials from the Starship? Mk1 was about 200t and they want it to be 120t. Also, what about the heatshield tiles? They will need some sort of structure/grid to hold them in place. That sort of thing will only add weight.

That's why I assume the payload to LEO to be about 100t or less. At the same time I'm fully aware that they have done magic to F9. It's insane how much that rocket evolved and how much its capabilities changed.

[Rakaydos]

2 hours ago, Wjolcz said:

Do you guys think they will be able to shear off 80t of materials from the Starship? Mk1 was about 200t and they want it to be 120t. Also, what about the heatshield tiles? They will need some sort of structure/grid to hold them in place. That sort of thing will only add weight.

That's why I assume the payload to LEO to be about 100t or less. At the same time I'm fully aware that they have done magic to F9. It's insane how much that rocket evolved and how much its capabilities changed.

Mark 1 was a battleship, overthick everywhere. They might not hit 120, but sub 150 is entirely reasonable.

The structure to hold the tiles is three rivits per tile. really not that heavy, all things considered.

[Wjolcz]

50 minutes ago, Rakaydos said:

The structure to hold the tiles is three rivits per tile. really not that heavy, all things considered.

Rivets sound like a bad idea though. Every single one of them seems like a good way to start a crack that will eventually rip the tank.

[kerbiloid]

  

3 hours ago, Wjolcz said:

Also, what about the heatshield tiles?

Spoiler

 

Edited January 3, 2020 by kerbiloid

[Rakaydos]

3 minutes ago, Wjolcz said:

Rivets sound like a bad idea though. Every single one of them seems like a good way to start a crack that will eventually rip the tank.

Works fine for submarines, doesn't it?

[Wjolcz]

1 hour ago, Rakaydos said:

Works fine for submarines, doesn't it?

The problem I see with this is that subs have a different structure. They aren't thin balloon tanks with rivets in them. They have multiple layers and structural support on the inside. If Starship was built like that it would probably have weighed twice of that (or more) what it already does (or rather did).

Edited January 3, 2020 by Wjolcz

[tater]

There are no rivets planned for Starship tile placement.

The test articles were nuts that were surface welded to the outside of the tank. The tiles were attached to the nuts. No perforations of the tank at all.

Edited January 3, 2020 by tater

[Wjolcz]

I should have probably googled the picture of the real thing before asking the question lol.

The attachment points are clearly visible. I'm guessing they already have a way to protect these attachment points.

[StrandedonEarth]

2 hours ago, Wjolcz said:

Rivets sound like a bad idea though. Every single one of them seems like a good way to start a crack that will eventually rip the tank.

Aside from the fact that it’s been said the tiles won’t be riveted, a hole will not start a crack, but a crack will continue to propagate once started. 

Fun fact: The shuttle was grounded for a few months in the 2000’s because cracks were discovered in the (liner of the?) main fuel feed lines inside at least one orbiter. The solution was to drill a hole at the end of the crack to stop it from propagating. I’m at work right now or I’d dig up a link. Maybe later. 

[sevenperforce]

9 minutes ago, Wjolcz said:

I should have probably googled the picture of the real thing before asking the question lol.

The attachment points are clearly visible. I'm guessing they already have a way to protect these attachment points.

They're not rivets; they're bolts. Presumably the nut will be welded onto the skin. The bolt-holes will be filled in with plugs.

[mikegarrison]

4 minutes ago, StrandedonEarth said:

The solution was to drill a hole at the end of the crack to stop it from propagating.

This is an extremely common means of stopping cracks in metal. A hole has much lower stress concentration than the sharp end of a crack.

[Elthy]

Arent airplanes using rivets a lot, too? Im not sure, but i guess that quality controlls are way easier for a rivet connection than for a weld...

[mikegarrison]

1 minute ago, Elthy said:

Arent airplanes using rivets a lot, too? Im not sure, but i guess that quality controlls are way easier for a rivet connection than for a weld...

Yes, this is historically one of the main reasons rivets have been used. These days there are more options available for reliable welding.

Another reason for using fasteners rather than welding is that fasteners don't disrupt the heat treatment of metal. If you see something like 2024-T3, the "2024" part describes the composition of the alloy and the "T3" part describes the heat and chemical treatment that was done to the alloy. High temperature welding undoes the heat treatment.

[Technical Ben]

On 1/2/2020 at 4:38 PM, kerbiloid said:

In that video they land the whole 2nd stage, while I mean separate the CrewDragon-derived propulsion part of it and land on ground like the CrewDragon original idea.
And spend the large cylindric tank part.

Just leave it in orbit, and use tanks to refuel/plug in... it's a tug. (Rendezvous with low orbit, then use it to boost up to the ISS and/or the moon. XD )

[StrandedonEarth]

2 hours ago, StrandedonEarth said:

The solution was to drill a hole at the end of the crack to stop it from propagating.

Odd, I could have sworn that was the solution. According to this article I was wrong. Maybe drilling holes was considered but ultimately not chosen. 

Not only that, but the cracks were theorized to start at holes (I’ll cover my butt by saying that I didn’t say a crack couldn’t start from a hole).But then, metal fatigue (from thermal expansion/contraction in this case) is a strange beast. 

Anyways, enough of this riveting tangential conversation for me

Edited January 3, 2020 by StrandedonEarth