totm nov 2023 SpaceX Discussion Thread

[tater]

6 minutes ago, Brotoro said:

The entry burn is relatively short, so they can't be counting on the deflecting effect of the exhaust to be the primary effect that keeps the stage from overheating...that only works during the burn.

Yes, this is certainly true. It could have an impact if it's sort of like MaxQ (move the bow shock away just before max reentry heating, plus slowing below that max.

Edited August 21, 2019 by tater

[magnemoe]

6 hours ago, StrandedonEarth said:

It’s been a bit of a chicken-and-egg thing. Re-use never caught on earlier because , as ESA argued, the number of payloads wouldn’t support it, as the factories (and workers) would be idled a lot of the time. But there weren’t a lot of payloads because launches were so expensive.

The USA basically priced themselves out of the commercial market, losing out to Ariane and Soyuz, with Uncle Sam being the sole remaining customer for American rockets. The American govt spent billions per year just to keep the Atlas factories open for when they were needed, at least at one point.

Then along came SpaceX, whose initial, expendable F9 1.0 was cheaper than almost anything else in the world, in its class. They’re slowing down now because they’ve caught up on their backlog, and will use their fleet of used rockets to launch Starlink, creating their own demand

Lots of issues, first it was an speed is primary objective cost does not matter up until 1970. The shuttle was an very capable craft but failed on cheap access to orbit. Yes part was mission creep, kind of trying to combine an submarine and an battleship, yes people did that with cruisers. No it did not work very well. 

 Second was how contracts in the US worked for an long time, you got your costs and an profit covered so it was no initiative to reduce cost. Number of payloads was also limited making reuse less relevant as you need an decent amount of launches to make reuse practical. 

Then you have the obvious trap of SSTO. Perfect is an enemy to good enough. 

Luck: Falcon 9 was very lucky as it has 9 engines so you could land on 1, you could not do that tricks with 2 or 4 engines. 
Adding wings and landing legs and land on an runway was an obvious approach but expensive to develop, see second. 
And yes the extremely kerbal development process, start with trying to crash stages on an position then start trying to destroy an barge.
No if you got politicians or bean counters in the loop that would not work. 
They say: Why are you destroying all the first stages? trying to inject they would be scrap anyway would be ignored. 

 

[magnemoe]

4 hours ago, mikegarrison said:

My best guess is that Falcon 9 launches provide a positive cash flow, and the company is using that cash to pay for these Starship and Starlink projects, which they hope will provide a good enough cash flow to pay for their next projects, ad infinitum. This is the same business model that most manufacturing companies have.

Starlink is likely to be the big golden goose. 
It has one major downside, bandwidth / km^2 is limited so its not an silver bullet in major cities but here you have lots of other options. 
In more rural areas or for ships and planes it an game changer.  Note that the starlink antennas is pretty expensive  but you could hook it up to an wifi network and cower your small town. 

No they will do this to maximize profit. Musk want to die as the emperor of Mars after all. 
Yes it fails, then I take Venus, its larger, harder to teraform but most of the infrastructure should be in place. 
You just need to expand it an high number of magnitudes, looking at an 286 chip.

 

[mikegarrison]

2 minutes ago, magnemoe said:

Starlink is likely to be the big golden goose.

Maaaaaybe. Lot of people have tried the same thing, and almost all of them have failed.

Typically one company spends all the money on the orbital infrastructure and then goes bankrupt. A new company buys the assets for nearly nothing and then operates the service. (cf. https://en.wikipedia.org/wiki/Iridium_Communications https://en.wikipedia.org/wiki/Teledesic https://en.wikipedia.org/wiki/Globalstar )

[magnemoe]

2 minutes ago, mikegarrison said:

Maaaaaybe. Lot of people have tried the same thing, and almost all of them have failed.

Typically one company spends all the money on the orbital infrastructure and then goes bankrupt. A new company buys the assets for nearly nothing and then operates the service. (cf. https://en.wikipedia.org/wiki/Iridium_Communications https://en.wikipedia.org/wiki/Teledesic https://en.wikipedia.org/wiki/Globalstar )

Add Iridium to the list they was almost deorbited then the US military needed all the satellite bandwidth at all cost asap because of war on terror. Premium for low weight devices, Think they bought them up then sold it. You don't want US military getting involved into search and rescue in Siberia after all. 
Its an matter of scale 10K satellites is more than 100. Stuff like phase array antennas was something who required an cruiser 30 years ago. Nowadays its used on radars on boats. 
 

[tater]

55 minutes ago, mikegarrison said:

Maaaaaybe. Lot of people have tried the same thing, and almost all of them have failed.

Yeah, as a fan of SpaceX I'm not sure I like betting the company on Starlink...

[mikegarrison]

4 minutes ago, tater said:

Yeah, as a fan of SpaceX I'm not sure I like betting the company on Starlink...

Are they really? Or is this just a side bet that would be nice to win but won't ruin them if they lose?

[Rakaydos]

4 hours ago, kerbiloid said:

Is it for sure that Space-X uses plain aluminium instead of waffles? It would be a back to 1950s. It sounds very strange.
Also their presentation shows some pattern (decorative or not) on the walls.

 

[tater]

34 minutes ago, mikegarrison said:

Are they really? Or is this just a side bet that would be nice to win but won't ruin them if they lose?

I guess it could be a side bet, it's not like it's at the expense of other launches. Also, the density of sats they can launch at once is... incredibly surprising.

When they first came out with this, I was looking at F9 and assuming they might be able to launch 20 at a time, and even with stage 1 reuse, it seemed expensive (100 F9 flights, min, or get SS working fast) when combined with the sats... It's looking like they are trying to be as cost effective as possible, so maybe it's not as risky as I first thought.

[mikegarrison]

"Waffels"? This is a very strange word to use for what you seem to be talking about (chem milling).

What strikes me is that relatively expensive processes like chem milling are actually more important in a rocket that will get used many times. The cost of doing it is the same whether you use the rocket once or 100 times, but if you use the rocket once you only get the benefit once, but if you use it 100 times you get the benefit 100 times.

[StrandedonEarth]

56 minutes ago, mikegarrison said:

"Waffels"? This is a very strange word to use for what you seem to be talking about

Yup, looks like aluminum waffles to me... (yes, it's Orion, but it was the easiest space-waffle image for me to find)

 

Edited August 22, 2019 by StrandedonEarth

[mikegarrison]

I doubt that subtractive manufacturing was used to make that structure you see in the Orion picture. That would be a *lot* of metal to remove compared to just adding frames and stringers. But maybe it was.

Poking around a little, some of the references do seem to indicate that big sections of that are milled from one piece of material. Interesting.

 

(I'm not a manufacturing engineer. I've always done aero/thermal/acoustic/combustion/etc. performance. I've often been surprised by finding out exactly how some of the parts are made.)

Edited August 22, 2019 by mikegarrison

[Nothalogh]

1 hour ago, mikegarrison said:

I doubt that subtractive manufacturing was used to make that structure you see in the Orion picture. That would be a *lot* of metal to remove compared to just adding frames and stringers. But maybe it was.

Poking around a little, some of the references do seem to indicate that big sections of that are milled from one piece of material. Interesting.

 

(I'm not a manufacturing engineer. I've always done aero/thermal/acoustic/combustion/etc. performance. I've often been surprised by finding out exactly how some of the parts are made.)

Since it's the pressure vessel, it would make sense that it be machined from a monolithic forging.

Likely mostly done with an absolutely enormous vertical boring mill.

[kerbiloid]

5 hours ago, magnemoe said:

Starlink is likely to be the big golden goose. 

Afaik, Space-X should get delivered first half of 12000 sats in 2023, otherwise the permission will be withdrawn.
Currently they have 60 of 6000. 3 years left.
6000 / 3 / 60 = ~30 launches per year just for Starlink.
Do they launch Falcons weekly?

3 hours ago, mikegarrison said:

"Waffels"? This is a very strange word to use for what you seem to be talking about (chem milling).

Sorry, "waffles". "A waffle envelope". Like this one.

Spoiler

I thought this term was adopted from English.

It makes an envelope 1.5..2 times lighter, while almost same strong, and this is a common practice since 1960s.
The holes are up to 9 times thinner than the edges.

44 minutes ago, Nothalogh said:

Since it's the pressure vessel, it would make sense that it be machined from a monolithic forging.

Any cylinder is made of connected bands.

Edited August 22, 2019 by kerbiloid

[mikegarrison]

Sorry if my typo confused you. Yes, I know the word "waffle" in reference to the food and occasionally to things that look like the food. I have never before heard anyone describe aerospace structures as waffles. Yes, I suppose I see the visual connection, but it's just not a word used in this context in my experience.

Generically, the term that covers this sort of structure is "semi-monocoque", although that's not used much in everyday discussion either.

[tater]

22 minutes ago, kerbiloid said:

Afaik, Space-X should get delivered first half of 12000 sats in 2023, otherwise the permission will be withdrawn.
Currently they have 60 of 6000. 3 years left.

Their license requires that they launch half of phase 1 by that date, which is actually only 2218 sats, I think. That's only 37 F9 launches at 60 per launch.

[Xd the great]

27 minutes ago, tater said:

Their license requires that they launch half of phase 1 by that date, which is actually only 2218 sats, I think. That's only 37 F9 launches at 60 per launch.

Aaand, if Starship works, they can cram 100 tonnes of starlink and launch it real quick.

[tater]

1 minute ago, Xd the great said:

Aaand, if Starship works, they can cram 100 tonnes of starlink and launch it real quick.

It's likely volume limited, so yeah, a lot.

[Kerbal7]

4 hours ago, kerbiloid said:

Afaik, Space-X should get delivered first half of 12000 sats in 2023, otherwise the permission will be withdrawn.
Currently they have 60 of 6000. 3 years left.
6000 / 3 / 60 = ~30 launches per year just for Starlink.
Do they launch Falcons weekly?

Sorry, "waffles". "A waffle envelope". Like this one.

  Reveal hidden contents

I thought this term was adopted from English.

It makes an envelope 1.5..2 times lighter, while almost same strong, and this is a common practice since 1960s.
The holes are up to 9 times thinner than the edges.

Any cylinder is made of connected bands.

Starlink needs to be cancelled. 5G is fast and cell towers are almost everywhere and soon to be worldwide. No one is going to be interested in laggy expensive internet. I don't like filling up the night's sky anymore with junk for my telescope anyways. This is another of Musk's money losers.

Like that silly Mars rocket he's building. It's nothing like a real spaceship and was constructed by a company that makes water towers. It's another of this charlatans PR stunts to hustle money out of investors. 

We do know Tesla is headed to bankruptcy as soon as investors get wise. It losing billions every year. And now with such bad management Musk is putting SpaceX in bankruptcy jeopardy too.

In five years, when the dust settles on Mar's rockets and Martian colonies we'll see what's left. Everything he touches loses money.     

[magnemoe]

10 minutes ago, Kerbal7 said:

Starlink needs to be cancelled. 5G is fast and cell towers are almost everywhere and soon to be worldwide. No one is going to be interested in laggy expensive internet. I don't like filling up the night's sky anymore with junk for my telescope anyways. This is another of Musk's money losers.

Like that silly Mars rocket he's building. It's nothing like a real spaceship and was constructed by a company that makes water towers. It's another of this charlatans PR stunts to hustle money out of investors. 

We do know Tesla is headed to bankruptcy as soon as investors get wise. It losing billions every year. And now with such bad management Musk is putting SpaceX in bankruptcy jeopardy too.

In five years, when the dust settles on Mar's rockets and Martian colonies we'll see what's left. Everything he touches loses money.     

The high speed as in 5G repeaters has an very limited range as in 150 meter.
You will hardly get any benefits outside 
5G serves the opposite purpose from starlink in that it provided bandwidth in crowded cities while starlink give bandwidth / km^2 so it works better who fewer people are around. 
And it probably become fast, a bit dependent on how routing works.

Finally 5G comes with the nice data plans from the telecom companies.  

[mikegarrison]

5G also has the side benefit of screwing up the accuracy of weather forecasting.

Oh, wait, sorry. That's not a benefit. Never mind.

[Technical Ben]

15 hours ago, tater said:

Why not both?

Supersonic retropropulsion works in that way (the exhaust moved the shock front away from the vehicle) (3 engines helps here, as I understand it). But during this process it also simply slows the vehicle such that there's less heating, anyway. Stage 1 booster is pretty slow, you have to remember, stage sep is at ~2.something km/s (usually low 2.something). So even a short 3 engine burn will strip a lot of that velocity off.

 

Yes, we know from talks/tweets that for RTLS (Return To Launch Site) they take about a 40% payload to LEO hit, and for ASDS landing, it's more like 18%. Improvements in the Merlins, as well as the move to superchilling has created so much margin that this is not a problem. It's not like anyone is going to fit 23 tons of cargo under a 5m fairing, really huge sats are less than 1/3 that mass, Falcon has loads of excess capability.

"Why not both?" because if it does not do both. It does not do both. If the retro grade burn/correction burn are not anywhere near the entry/heating/atmospheric conditions, then *it does not do atmospheric deflection*. The question is "does it", not "can it".

[kerbiloid]

https://forum.nasaspaceflight.com/index.php?topic=39578.0

 

[tater]

1 hour ago, Technical Ben said:

"Why not both?" because if it does not do both. It does not do both. If the retro grade burn/correction burn are not anywhere near the entry/heating/atmospheric conditions, then *it does not do atmospheric deflection*. The question is "does it", not "can it".

Yeah, but I think that short of asking the question to someone at SpaceX and getting a definitive answer, we're not going to know the relative balance of the 2 effects for sure, and have to make some educated guesses.

Merlin 1D can throttle, but not super low, I think min throttle is on the order of 40%. The entry burn is typically about 20 seconds, but we are unsure of the throttle setting for the 3 engines used. Boostback is also a 3 engine burn (gotta be full thrust I'd think), and it's like 30 seconds, and obviously scrubs off all the downrange velocity, then imparts velocity in a vector for RTLS, so depending on the throttle a 20 second burn could either scrub off all the velocity, or at least a large % of it. I tend to think that it's scrubbing off most of the velocity propulsively but there must also be a component of bow shock displacement in terms of reduced heating load.

[Technical Ben]

1 hour ago, tater said:

Yeah, but I think that short of asking the question to someone at SpaceX and getting a definitive answer, we're not going to know the relative balance of the 2 effects for sure, and have to make some educated guesses.

Merlin 1D can throttle, but not super low, I think min throttle is on the order of 40%. The entry burn is typically about 20 seconds, but we are unsure of the throttle setting for the 3 engines used. Boostback is also a 3 engine burn (gotta be full thrust I'd think), and it's like 30 seconds, and obviously scrubs off all the downrange velocity, then imparts velocity in a vector for RTLS, so depending on the throttle a 20 second burn could either scrub off all the velocity, or at least a large % of it. I tend to think that it's scrubbing off most of the velocity propulsively but there must also be a component of bow shock displacement in terms of reduced heating load.

No. There will not be. Look at where it is in its trajectory and entry. Thus if it's not there, it won't be there.