totm nov 2023 SpaceX Discussion Thread

[tater]

I think it's not comparable in terms of cycles. The pressures are lower, the temperature extremes are lower, and it's not cryogenics.

It's one thing to make a launch vehicle out of composites. It's another to make one to make a spacecraft capable of multiple reentries and reflights, capable of holding LOX literally for years.

I think the booster is non-trivial, but straightforward. The unmanned spacecraft is harder, but likely possible, but I think the reuse of the spacecraft is the really tricky bit in terms of the composites.

[DerekL1963]

1 hour ago, tater said:

I think it's not comparable in terms of cycles. The pressures are lower, the temperature extremes are lower, and it's not cryogenics.

Pretty much this.  787 cycles more but the extremes are less.  The temperature cycling is over a much narrower range (and are not cryogenic).  Etc... etc...  These differences matter.  BFR is a huge leap outside of current operational experience.

That leads the cautious and experienced to be... a little less fanatical and a lot less trusting in extrapolations.
 

3 hours ago, CatastrophicFailure said:

They’ve been made big before, they’ve been made rockets before, and they’ve been made high-pressure cryogenic vessels before. What SpaceX is doing is combining it all into one. 

In other words, they're doing something never done before.
 

3 hours ago, CatastrophicFailure said:

Remember, they have already made a full-diameter test tank, taken it to flight pressure/temperature, and then deliberately tested it to failure. They didn’t seem to have any insurmountable problems. 

SpaceX isn't in a habit of discussing their problems.  Quite the opposite in fact, unless the problem occurs in the public eye they don't discuss it at all.

[CatastrophicFailure]

21 minutes ago, DerekL1963 said:

In other words, they're doing something never done before.

What they're doing is evolutionary, not revolutionary, like all their other advances. Standing on the shoulders of giants, to borrow the phrase.. They aren't groping around blindly with unknown materials in untested applications. Recognizing this is not "fanatical" , it's looking at the big picture... with a dose of optimism. You can be a skeptic if you want, that's fine. Optimism is much more fun. 

24 minutes ago, DerekL1963 said:

SpaceX isn't in a habit of discussing their problems.  Quite the opposite in fact, unless the problem occurs in the public eye they don't discuss it at all.

They also aren't exactly Gradatim Ferociter-ing, like with Falcon Heavy, are they? What with their big expensive tool in a makeshift tent at an unbuilt factory, no doubt with more in the pipe... this speaks to me that they have confidence in that they're doing and that the development is more-or-less proceeding as expected. ElonTime notwithstanding.

[magnemoe]

8 hours ago, tater said:

I think I need to reiterate the importance of Starlink. 4,425 sats. It has the possibility to make vast quantities of money, as well.

Let’s assume Musk’s Mars goals are his real aim, and that it is achievable in some fashion. It will still take money, and lots of it. Contrary to fanboi myth, there is no possible economic driver for Mars. It’s a sinkhole for cash. If he wants to do it, he needs Starlink money. He’s bet the whole sat constellation on flying BFR soon enough to fly at least half of those in 6 years. BFR is the only way that ever happens.

He has to launch 2,213 satelites of 400 kg,  that is 885 ton, or 80 launches of falcon 9 assuming 10 ton payload capacity. Around 13 launches each year. 
pretty plausible.
Assuming they will use aging first stages here on many launches as you can take some fails 

yes if they get BFR up and running in time they will use that. again you can take far higher risk with mass produced satellites. 
But they don't have an pad for launcing BFR into polar orbits as I know. 

[Wjolcz]

9 hours ago, tater said:

I think I need to reiterate the importance of Starlink. 4,425 sats. It has the possibility to make vast quantities of money, as well.

Let’s assume Musk’s Mars goals are his real aim, and that it is achievable in some fashion. It will still take money, and lots of it. Contrary to fanboi myth, there is no possible economic driver for Mars. It’s a sinkhole for cash. If he wants to do it, he needs Starlink money. He’s bet the whole sat constellation on flying BFR soon enough to fly at least half of those in 6 years. BFR is the only way that ever happens.

My thoughts exactly when I first heard of Starlink. The commercial BFR (Earth-to-Earth) never seemed like a good way of financing the whole Mars plan. It would be cool to have such thing but there are so many risks and legal issues that I don't think rocket commercial flights will become a thing in the near future. A lot would have to change.

[Wjolcz]

1 hour ago, magnemoe said:

But they don't have an pad for launcing BFR into polar orbits as I know. 

I thought it will be launching from water? A launchpad on the sea gives a lot more capabilities. You can move it closer to the equator or farther from land. If I'm not mistaken they could just build it on the west coast, put it on a boat, go west far enough and still launch east. Stormy weather is still a problem though. They would probably need an oil rig-like launch pad. If they are building a factory on water anyway then what is stopping them to have a part of it detachable? You build the rocket on water, put it on the pad, haul the pad wherever you please, launch.

I think I have talked about this already but on reddit?

[tater]

2 hours ago, magnemoe said:

He has to launch 2,213 satelites of 400 kg,  that is 885 ton, or 80 launches of falcon 9 assuming 10 ton payload capacity. Around 13 launches each year. 
pretty plausible.
Assuming they will use aging first stages here on many launches as you can take some fails 

They have launch 2 small testbeds, not real StarLinks. Even if they mass 400kg, they could very well be volume limited. Iridium-NEXT are what, 800kg? They put 10 in Falcon, right? Call it 20 Starlinks/launch. That's 111 launches. Twice that if only 10 fit, or somewhere in between.

 

[Ultimate Steve]

I heard on Reddit or something (it had a source that was pretty official) that SpaceX were planning on launching 25 Starlink sats per F9 launch.

[tater]

3 minutes ago, Ultimate Steve said:

I heard on Reddit or something (it had a source that was pretty official) that SpaceX were planning on launching 25 Starlink sats per F9 launch.

Gotcha I hadn't heard the number.

The 2213 is the minimum they need to launch, however. I'm unsure if they want to take 10 years to complete the constellation from a business standpoint. Assuming they want to have it operational sooner, it would reasonable to double that number of regular, dedicated launches.

[PB666]

8 minutes ago, tater said:

They have launch 2 small testbeds, not real StarLinks. Even if they mass 400kg, they could very well be volume limited. Iridium-NEXT are what, 800kg? They put 10 in Falcon, right? Call it 20 Starlinks/launch. That's 111 launches. Twice that if only 10 fit, or somewhere in between.

 

The satellites have different orbits. For example you would put equitorial and polar orbiting satellites on the same launch. 

Cape canaveral affords a 28.3' inclination, but the ascending node rotates along a 360 degree axis, so that if you wait seperate launches by T*2pi/24h - n2pi/24h you can have two launches with orbital inclinations of 56.3' relative inclination. Indeed this is a minimum, for leo communication satellites you may want to leave at a 45' bearing at launch to cover more of the globes surface at close range.

The assumption here is this. For any orbit you want many satellites, but with any orbit you can station satellites by slighlty altering apogees and then reducing apogee when its perigee position approximates its station. 

But for an immediately functioning satellite network filling all the stations of one orbit makes no sense, because then your network provides access twice a day for a brief period of time. Its somewhat more intelligent to stagger the ascending nodes of the launch e.g 0, 180, 90, 270, 45, 135, 225, 315, ..........

9 hours ago, tater said:

I have no doubt it can work for some time period. I think the question is one of cycling, particularly once the heat loads on it become more substantial. The principle concern with composites is delamination as I understand it. Holding together for a few hours is one thing, months and years is another.

If nothing happens but a BFingRocket flying---that's interesting, too!

Well, that would be a problem with the launch phase. The orbiter would be exposed to space weather but the tanks would not be. The extremes of cryogenics can be partially eliminated simply by placing a two-stage refridgerant line into the tank and constantly keeping the tank cold. 

I think that if you get 10 launches out of the launch phase the marginal gain utility of the next 10 launches drops and the trend continues as such that only if the market is extremely competitive do you want to risk failure by extending the launches. 

[PB666]

10 hours ago, DerekL1963 said:

787 fuseleages don't contain cryogenic fuels nor do they routinely experience the stresses that a rocket fuselage does.  Yes, he is reinventing the wheel.

What about lithium ion battery explosions in the avionics bay?

6 hours ago, CatastrophicFailure said:

What they're doing is evolutionary, not revolutionary, like all their other advances. Standing on the shoulders of giants, to borrow the phrase.. They aren't groping around blindly with unknown materials in untested applications. Recognizing this is not "fanatical" , it's looking at the big picture... with a dose of optimism. You can be a skeptic if you want, that's fine. Optimism is much more fun. 

They also aren't exactly Gradatim Ferociter-ing, like with Falcon Heavy, are they? What with their big expensive tool in a makeshift tent at an unbuilt factory, no doubt with more in the pipe... this speaks to me that they have confidence in that they're doing and that the development is more-or-less proceeding as expected. ElonTime notwithstanding.

Don't minimize the scientific prowess of SpaceX they are already the leading sxpert on atmosperic reentry and propsive landings. This assumption is that there is no science in the engineering, thats a false assumption.  I should point out one other thing; if they are really looking at shockwave and material stress, your focus should be on the highest speed bullet trains, as for example going through a tunnels whose walls are half a meter from the trains skin at half the speed of sound and doing this every 5 minutes. Changes of pressure steep higher are much more frequent. 

[wumpus]

48 minutes ago, PB666 said:

The satellites have different orbits. For example you would put equitorial and polar orbiting satellites on the same launch.

Starlink plans to have 83 specific planes (inclinations) used.  While some will be more full than others, with ~2000-4000 satellites total it seems that even a BFR wouldn't fill a single plane in a launch.  Inclination changes would presumably be a single burn to insert all birds into the same plane.

I'm curious to see how much mass a block 5 FH can lift to LEO (with arbitrary inclination changes).  The current FH can't lift much beyond ~68 tons to anywhere (limited by structural support), but there was obviously a chance to change that in block 5.  I suspect that Elon Musk didn't hedge his bets at all and that this is purely a job for BFR.

For those wondering about the business: 200,000 people coughing up $50 a month (the only figure I've seen, but it certainly would be competitive in the US.  Developed nations might have cheaper ISPs that are harder to undercut) for 10 years will provide a billion dollars (ignoring interest.  That will be a large chunk of any calculations considering sunk BFR costs and the four year launch plan).  I'm sure there are significant non-infrastructure costs, but still that has to be the biggest cost of an ISP and the biggest barrier to competition.  We also don't know the NRE and production costs of the satellites.  I sure nobody has ever tried to build 4000 copies of one spacecraft.

11 hours ago, tater said:

Boeing built a 5.5 m tank for NASA, actually, and the idea has been worked on at NASA for years now.

Why do I get the feeling that Spacex is all over this research and that ULA won't use it without additional NASA/DoD/US govt funds specifically earmarked for a composite tank?

Edited April 10, 2018 by wumpus

[tater]

Assume the 80 dedicated launches above. What does an upper stage cost? 10 M$? That's 1.77 B$ in upper stage cost for the constellation. Perhaps the goal is to deploy as many as possible using a reusable upper stage?

[sh1pman]

36 minutes ago, tater said:

reusable upper stage?

You mean BFR?

[tater]

38 minutes ago, sh1pman said:

You mean BFR?

Yes. BFS (cargo). Meaning they don't want the additional 400k per sat in lost Stage 2 costs. If a BFS costs a couple hundred million to make, the cost of 80 stage 2s could build perhaps 9 BFS (or some complete BFR/BFS stacks, and the facilities to launch them)

[PB666]

1 hour ago, tater said:

Assume the 80 dedicated launches above. What does an upper stage cost? 10 M$? That's 1.77 B$ in upper stage cost for the constellation. Perhaps the goal is to deploy as many as possible using a reusable upper stage?

You could theoretically make several small plane changes with one BFR upper stage. And the problem here economically is that the more sateelites per launch, the S2 costs start becoming trivial. Since there are going to be competitors you whats going to get as many functional satellites as possible as a driving force for business growth. Again planting all satellites on a single plane or few close planes is not beneficial to the demand side of the business. 

[tater]

4 minutes ago, PB666 said:

You could theoretically make several small plane changes with one BFR upper stage. And the problem here economically is that the more sateelites per launch, the S2 costs start becoming trivial. Since there are going to be competitors you whats going to get as many functional satellites as possible as a driving force for business growth. Again planting all satellites on a single plane or few close planes is not beneficial to the demand side of the business. 

Yeah, this is true. Of course they might have propulsion on their own (ion) to slowly plane change.

 

[DerekL1963]

6 hours ago, magnemoe said:

He has to launch 2,213 satelites of 400 kg,  that is 885 ton, or 80 launches of falcon 9 assuming 10 ton payload capacity. Around 13 launches each year. 
pretty plausible.

Presuming SpaceX can produce second stages fast enough, and presuming they can refurbish first stages fast enough/have enough to throw away, and presuming the range is available - and all this on top of their existing manifest.
 

2 hours ago, PB666 said:

Don't minimize the scientific prowess of SpaceX they are already the leading sxpert on atmosperic reentry and propsive landings.   This assumption is that there is no science in the engineering, thats a false assumption.

Which sure sounds impressive taken as a sound bite.  But, once again, broken down into reality - you're talking a couple of dozen entries and landings.  The science of engineering says that's a very narrow base of experience from which to extrapolate to a much larger vehicle.  Worse yet for your nonsensical claim, the experience base was much narrower when they announced the BFR last year and not much bigger when they rolled out the test article.

Or, to put it another way, I am minimizing nothing and assuming nothing.  I'm proceeding forward from the facts and taking the actual science into account.  Science and engineering aren't buzzwords or sound bites, they're very real things.
 

3 hours ago, PB666 said:

 I should point out one other thing; if they are really looking at shockwave and material stress, your focus should be on the highest speed bullet trains, as for example going through a tunnels whose walls are half a meter from the trains skin at half the speed of sound and doing this every 5 minutes. Changes of pressure steep higher are much more frequent. 

"Half the speed of sounds" sure sounds impressive.  But the shockwaves we're talking about occur in the supersonic range, and the stress comes from sustained acceleration and vibration quite unlike that which Shinkansen experiences.  To add to those pesky scientific and engineering facts we also have to consider that a Shinkansen doesn't have be as lightweight as a booster.  (Not to mention that they don't "go through a tunnel every five minutes".)

[tater]

2 minutes ago, DerekL1963 said:

and presuming the range is available - and all this on top of their existing manifest.

Wow, this. Even granting them all the rest in terms of cadence, the range time itself is a huge issue. Forget their own delays, even, and just consider normal delays like weather, wayward boats, etc. Adding even a single launch a month on top of other customers would be really difficult.

[YNM]

5 minutes ago, DerekL1963 said:

Presuming SpaceX can produce second stages fast enough,

It's reusable.

It's a glorified Space Shuttle.

[tater]

Just now, YNM said:

It's reusable.

It's a glorified Space Shuttle.

He's replying about the idea of launching the constellation with the existing Falcon 9 system, not BFR.

It's not impossible, I suppose, but that idea seems pretty difficult, particularly if the goal is to move many production people to the next gen vehicle, and only keep people on F9 who can get them ready to refly (S1, and fairing), and build new second stages.

[YNM]

4 minutes ago, tater said:

He's replying about the idea of launching the constellation with the existing Falcon 9 system, not BFR.

Ah.

In which case, I agree with @DerekL1963 that it's stupendously unfeasible.

 

But then we're back to the glorified Space Shuttle...

[tater]

1 hour ago, YNM said:

But then we're back to the glorified Space Shuttle...

Well, at least the current information shows that they are actively working on it in some earnest for testing.

Clearly a lot will depend on testing, and I fully expect them to break some of them in the process, lol. What SpaceX is trying here is as @DerekL1963 points out, not a simple sidestep from existing aircraft manufacturing, or even a slight evolutionary change from that. It's certainly mainstream to the extent RocketLab is doing it (with an n of 1 or 2 at this point, lol), but it's a substantially different goal to make a fully reusable craft than one that when it comes down to it literally only has to survive a few hours (most of which are on the ground).

 

Boeing 5.5m tank example vid:

 

Edited April 10, 2018 by tater

[PB666]

1 hour ago, DerekL1963 said:

Presuming SpaceX can produce second stages fast enough, and presuming they can refurbish first stages fast enough/have enough to throw away, and presuming the range is available - and all this on top of their existing manifest.
 

Which sure sounds impressive taken as a sound bite.  But, once again, broken down into reality - you're talking a couple of dozen entries and landings.  The science of engineering says that's a very narrow base of experience from which to extrapolate to a much larger vehicle.  Worse yet for your nonsensical claim, the experience base was much narrower when they announced the BFR last year and not much bigger when they rolled out the test article.

Or, to put it another way, I am minimizing nothing and assuming nothing.  I'm proceeding forward from the facts and taking the actual science into account.  Science and engineering aren't buzzwords or sound bites, they're very real things.
 

"Half the speed of sounds" sure sounds impressive.  But the shockwaves we're talking about occur in the supersonic range, and the stress comes from sustained acceleration and vibration quite unlike that which Shinkansen experiences.  To add to those pesky scientific and engineering facts we also have to consider that a Shinkansen doesn't have be as lightweight as a booster.  (Not to mention that they don't "go through a tunnel every five minutes".)

The vibration and stretch forces are quite intense, if you have ridden on these bullet trains you would know this. A 787 only experiences stretching forces relative to ambient along the length, a bullet train experiences both compression and inflation forces and the changes occur very rapidly measured in millesonds. Its not an issue of how much force, but given the materials you decided to use and the number of times and how rapidly they are stretched. . . . and via stress testing on the material shows how much more strength needs to be added. An excellent example is a centrifuge rotor. At the beginning of its life you can spin at maximum RPM. Howver the faster you spin it its ages as a function of the square of the speed. At some point the rotir begins to strech and becomes unbalanced. If you never spin the rotor below 85% of its rated RPM then it can be used until the materials themselves (i.e corrosion) degrade. But the rotor ages with expansion contraction events. A rocket is likely never to go through MaxQ 100 times or a 1000 times. A bullet train may reach its extremes 100 times a day. 

The point about SpaceX is that they have access to NASAs science. In addition they are willing to experiment with various settings. Again, who has done 1000s of launches? They may not have the data, but they clarly have enough engineers and scientist to figure it out. Again these are for decisions that can be made in the operations around Earth. As we know they have used rockets that can be easily used for any type of testing, including sounding rockets. 

The problem regards testing with regard to Mars and details of Mars that are nit easily reproduced on Earth, when you subject is 3 to 21 months away it does not easily accessed, then it means more testing on Earth is required. 

2 hours ago, tater said:

Wow, this. Even granting them all the rest in terms of cadence, the range time itself is a huge issue. Forget their own delays, even, and just consider normal delays like weather, wayward boats, etc. Adding even a single launch a month on top of other customers would be really difficult.

They need to be able to make several launches in the range. 

[tater]

35 minutes ago, PB666 said:

They need to be able to make several launches in the range. 

Range congestion is already an issue, and once Blue starts flying from the Cape as well, it can only get worse. If SpaceX never exceeds their pace for this year (~30), and Blue flies what they claim (12/yr), then the range already exceeds what the AF says they can possibly do in a year (48).