Building A Starship In Texas
#Starship #SpaceX
The orbital version of Starship is now under construction at @SpaceX in Boca Chica Beach, Texas! ?? yeah!! pic.twitter.com/Cr93kA4ICo— Evelyn Janeidy Arevalo (@JaneidyEve) April 7, 2019
#Starship #SpaceX
The orbital version of Starship is now under construction at @SpaceX in Boca Chica Beach, Texas! ?? yeah!! pic.twitter.com/Cr93kA4ICo— Evelyn Janeidy Arevalo (@JaneidyEve) April 7, 2019
just based on the pix, that looks like a Highly uncontrolled assembly. Could this be a mock up, rather than flight h/ware ?
Who knows what lurks in the hearts and minds of men? Elon Musk (aka “The Shadow”…from the old radio mystery series) knows! Mmmhahahaha!
I do not think Elon Musk would go to that much trouble for a life size mock up. Possible first orbital flight next year is very ambitious so I would guess this is the real thing.Remember, he does things differently, such as make Teslas under a tent.
I still think falcon heavy with a dragon 3 lunar and a dragon 4 maritan is what will happen next year, and the year after. Lunar version with custom lunar support stage for dragon 3 lunar and Martian version with LEO docking with multistage custom return and life support stages.
https://forums.teslarati.co…
No chance. SpaceX is all in on Starship/Super Booster like Boeing was all in on B-747 development.
It’s stainless steel! They could put it together in the middle of a hurricane raging!
Oh god, not this again.
I’m thinking of the specialized gear and tooling over at Michoud where, it is said, that tolerances approach the second or even third decimal place. It’s impressive as hell, even when leveling is problematic.
Here we have some guys slapping together panels together over (presumably) some sort of form. Not sure what I am missing here.
Possibly margin and efficiency. Those extra significant figures are often there to make something more efficient. You often don’t need them if you can afford to say, “I’ll just round up and it will definitely be good enough.” But that does hurt performance and (mass) efficiency. People have argued that NASA and vehicles like SLS take things too far in the direction of optimizing for performance. Bu it’s also quite possible that SpaceX and Starship are taking things too far in the other direction. We’ll find out when it flies.
certainly dim’ns could be over defined (unnecessarily tight) in the search for perfection. Another take is tightly controlling dim’ns makes for a high quality component. It does seem “bush league” to be building a rocket this way.
As I said, you can go too far in either direction. It looks like NASA has gone too far in one direction. SpaceX may well be going too far in the other direction. But that remains to be seen. If they have, then they will have wasted a whole lot of their own money. But if it works, that would be revolutionary.
The beauty of this idea is that SpaceX wouldn’t be wasting a lot of money if this flops, they’re just paying a dozen or so welders to weld up steel sheets from vendors, outside a tent. If it doesn’t work, they wasted some time, but the cost to them is very low.
For assembly yes but the engineering work is still significant, the Raptors apparently required some redesign after the first brief tests at Boca Chica.
What I saw on Twitter was that the Raptor was removed for inspections. I didn’t hear anything about a “redesign after the first brief tests at Boca Chica.”
The “redesign” likely comes from a test to near-destruction test done using Raptor SN01. After spinning it up to ? levels it was examined to inform the SN02 and subsequent engine builds. SN02 was used for tests and is being examined. SN03 is on the stand at McGregor.
I suspect they will erect an assembly building at some point before they go to commercial operations, if only to protect against another windstorm. However just the fact that the tank does not leak suggests that it was put together pretty well from the point of view of welding and plumbing. However I think SX was wrong to scrap the finer winding rig, if it wasn’t useful they could at least have sold it to Blue Origin, which are going with fiber.
When I hear ‘bush league’ a certain dropped tank dome with no backup comes to mind, along with photos of all sorts of parts over at SX.
I have read that it is assembled of quarter inch steel plates. That seems a little heavy for an actual spacecraft. The Atlas I and II were pressure-supported structures built from 301 stainless steel (16-18% Cr, 6-8%Ni) in “extra-full-hard” grade in a thickness of only a couple of millimeters, so thin you can indent it slightly with your thumb, and its corrosion resistance here in Florida is remarkable; the Atlases in the display settings are still corrosion free after decades in salt air. The craftsmanship demonstrated by butt welding of what is essentially sheet metal is outstanding.
Maybe the dimensional tolerances are a little loose on the Starship prototypes, but if the welding is close to the quality that was needed for the Atlas they might just fly.
You are not thinking this through. Those early Atlas with the stainless steel balloon tanks are not design to reenter the atmosphere and propulsively land on their tails like RAH intended. As all other current launchers.
Whether a balloon tank is useful is a matter of whether stiffness or strength is limiting. Internal pressure will not add strength but it is a very light way of adding stiffness. Even at a quarter inch thickness I suspect the Starship with its large hull diameter will need some internal tank pressure to prevent buckling during entry.
The reduction in weight possible with the pressure supported structure was such that the entire Atlas booster (minus two of the three engines which were dropped) could be placed in orbit, and was on at least one occasion. It was the closest to a single stage to orbit vehicle we have had. I agree the Atlas was not designed to withstand entry stress.
I have little doubt it will fly, just a question of how fast and how high 😉 We don’t know the particular objectives they hope to achieve with the new test article, so the mass and structural requirements are unclear. For comparison, apparently the tanks of the Saturn V first stage were 60 mm thick aluminum, see:
https://history.nasa.gov/SP…
so assuming 1/4″ stainless steel is correct, that’s 1/10 the thickness of the Saturn V tank. Unsure if the shell seen doubles as a tank wall, so may not be a direct comparison, but interesting…
Clearly the Saturn tank wall was not 6 cm thick. The reference is ambiguous but it is possible the tank walls were machined with integral stiffeners from 6cm plate, leaving the skin thin in most areas and the stiffeners extending inward from it.
I wouldn’t have thought so (which is why I said “apparently” and gave the link), but it’s been about 40 years since I did structural analysis as an undergrad so I don’t have a feel for it. Probably a typo or mis-worded. From another MSFC Saturn V page, the fuel tank weighed 12 tons empty and was 43’ tall. Approximating the shape as a cylinder, and ignoring the internal baffles, that gives a tank wall thickness of about 7 mm. So perhaps it was a typo and should have been 6.0 mm rather than 60 mm… Which is about the thickness you mentioned for the hopper steel plate.
The Saturn was about 1/6 the weight of the Starhopper tank, due in part to the aluminum construction. There’s no magic at SpaceX that makes weight unimportant. I’m not saying it cannot be done, but the business risk is high and economics is ruthless. Where is the heavy payload market? The New Glenn will have the same capacity as the Falcon Heavy (70 tons to LEO) and a lower operating cost. The FH launching tonight doesn’t even have a payload that requires its performance, and most Ariane V launches were dual payload. I would have suggested a single hull LV in the FH class using the Raptor engines first.
Judging by Falcon, as you add reusability, the price drops. So a partially-reusable heavy is cheaper than an expendable medium. A fully-reusable super-heavy may be cheaper to launch than a small-sat launcher.
If SpaceX gets within a factor of 5 of its stated launch costs for Starship (and it’s been much better at predicting capacity and price than it is sticking to schedules.), it will destroy the rest of the launch market. Even small-sat launchers, IMO.
Irrelevant. Even if BO starts moving faster than at any other point in 19 year existence, It will only come into service when Starship is approaching operation and Falcon is losing customers who are willing to wait for a lower-cost flight on Starship.
We don’t know how well BO will handle costs. So there’s no, zero, numbers to back that up.
In triple-core recovery mode, GTI+1500, it does.
But even if it doesn’t, it makes my point. The client preferred the lower price of FH-reusable than F9-expendable.
And the clients weren’t really happy with that. Which is why the newer Ariane is going to be smaller.
That was SpaceX’s plan before Falcon 1 even launched. Falcon X with Merlin 2 engines.
Falcon 1 – Single Merlin 1 engine.
Falcon 5 – Five Merlin 1 engines.
Falcon 9 – Nine Merlin 1 engines.
Falcon 9 Heavy – Triple F9-core.
F9 v2 – With a single Merlin 2 engines.
F9H v2 – Three cores, hence three M2 engines.
Falcon X – three M2 engines on a single core.
Falcon X Heavy – Triple core Falcon X.
Falcon XX – Nine M2 engines, single core.
Plans changed. Falcon 5 was just skipped over. Merlin 2 was meant to be kerolox on the scale of F1, as a plug’n’play replacement for 9xM1s, so switching to methalox makes it wasteful trying to convert Falcon. SpaceX is much more comfortable flying multi-engined cores, so there’s no real pressure to enlarge Raptor as much as Merlin 2 was intended. And Falcon Heavy convinced Musk that a larger single core is better than triple-core. (Although once FH is better proven, he might do another 180 and decide to triple-core BFR.)
A triple core Super Heavy booster would have 93 engines and over 36 million pounds of thrust at launch. Since the Saturn V with 7.5 million pounds of thrust was knocking down press booths 4.5 kilometers away (watch the Apollo 4 launch video) that Beast would need a 15 kilometer safety zone when it launched.
Atlas was a 1.5 staging vehicle, pretty close to SSTO, it only drops two engines during ascent. Starship is 2 stages, so more tolerant to weight growth.
Quarter inch is 6.35mm, Starship height is 55m, diameter 9m, so tank skin volume is 9.9 m^3. Stainless steel density is 8 ton/m^3, so tank weight is about 80 tons. Assuming tank weight is 60% of total weight, this gives total weight of 133 ton for the prototype. Originally BFS dry weight is supposed to be 85 tons, so it’s about 50 tons over weight. Given its payload capability is over 100 tons, a 50 ton overweight is not too concerning for the first prototype.
{laughs}
“Our rocket is 50 tonnes overweight! Meh. Still works.”
Only SpaceX.
“slapping together panels together”
reminds me of high precision machining and heavy armor of a WWII Tiger tank vs a Sherman tank. Sherman is no match against a Tiger but can be mass produced and faster servicing. Most tank uses do not battle other tanks.
Exactly! Space X, like the Sherman, is good enough for expanding our frontiers. NASA, like the Tiger Tank, may be better in some respects, but is not suitable for expanding our frontiers as evidenced by the fact that humans have not been out of low earth orbit since 1973.
If you want a analogy to World War Two vehicles, the Zero (A6M) aircraft might be a better one. Horikoshi and his team at Mitsubishi really pulled out the stops in terms of cutting down weight to improve performance. Design tolerances which pared down excess structural weight, new and then-exotic aluminum alloys, etc. That made it (arguably) the highest performance fighter aircraft at the time in was introduced in 1939. It also made it quite fragile and easily damaged when hit. High performance but not exactly study or robust. I think that’s a fair description of conventional launch vehicles compared to one made for reuse and out of welded steel.
Same with the Me-109 vs the Mustangs.
The 109’s engine had to be shipped back to the factory for repairs. The Merlin engine (hey!) could be field-repaired by any competent engine mechanic.
The 109 was superior when it worked. The Mustang worked.
SX probably bought a precision sheet metal roller or five for the cost of daily break time on the SLS.
Musk is probably thinking that extra 1% or 2% capability isn’t work the extra $5 billion in production costs.
And if it turns out to be 20% or 50%, the whole idea could be a flop. But trying out wild ideas to find out is the best way to make progress.
Certainly agree with that. I assume Space X engineers ran the numbers so the risk is worth taking.
What’s insane is that Starship can be 50% undercapacity and 300% overprice and it’ll still destroy anything else currently on the market… or planned, and not per-kg but per-launch.
However, given that SpaceX has a pretty good record for exceeding predicted payload capacity, and not exceeding 200% of predicted price, it seems unlikely that the enormous margins will prove necessary for Starship to be a commercial success.
I’d rather wait and see about the SpaceX Starship. I don’t disagree with what you wrote. But the inital version of the Falcon 9 was much less capable than the current block 5, was not capable of landing or reuse, and it’s first launch was way behind the original schedule.
I think the results were fantastic, and even if the Starship is an order of magnitude less capable then SpaceX claims it would still be revolutionary. I’m just reluctant to bet the farm on everything working as well as they say. Or even a hundredth as well as they say. I’m a fan of the Falcon because they actually made it work. Even if I’m not from the “show me” state, I still want to see Starship fly before I am confident about the claims.
SpaceX’s scheduling I’m skeptical of. Absolutely. But I’m no longer doubtful that they will deliver something that meets their goals. Their record is sufficiently deep now to set aside skepticism about capability.
However, it met (and exceeded) it’s originally stated target payload capacity of 11 tonnes to LEO.
Likewise, reusability wasn’t a stated goal for F9 when the initial plan was released. Instead the goal was to keep launch costs below $50m. Actual price was around $65m by the time it flew. FH has similarly exceeded its predicted payload capability, in spite of losing cross-feed.
Re: Schedules.
I’ve noticed that If you plot SpaceX’s predicted schedules as if they were Mars years, then map them back onto Earth years, they fit surprisingly well. Adding further evidence to the theory that Musk was originally from Mars, not Sth Africa.
No, the real Martians are emigrated from Hungary, mostly from Vienna. That was a running joke about the large number of really smart mathematicians and physicists who emigrated from there in the late 1920s and 1930s. (Teller, Van Neumann, Van Karman, Szilard et al.) Claiming to be emigrants from a foreign country would explain their accent; claiming to be from the same country would make forging documents and records easier; claiming to be from Hungary would hide the fact they spoke their allegedly-native language with an accent (since virtually no one outside of Hungary speaks Magyar well enough to notice that); etc.
I continue to believe that these are hoppers only and will not be going into orbit, or even all that high in the atmosphere. Thus the production quality we see on these vehicles is not indicative of the final product. Nor is the production location, which I think is currently in Boca Chica only for expediency as related to these particular tests. No different in concept to the Grasshopper flights in McGregor, other than it was easier to build the Grasshoppers in California and ship them to Texas, whereas due to its massive size it is quicker and easier to build the Starship hoppers onsite. They may also not plan for these to go very high or fast, having learned from Grasshopper which was originally planned for later high altitude test flights (higher than would have been allowed in McGregor) but then they decided that they got enough data from the low altitude tests and went straight from there to orbital test flights (using different hardware). So I am guessing a similar approach will be used with Starship and thus the tolerances don’t need to be as tight for the hoppers. Not that they will ever go to the precision of Michoud, which is the other extreme as fcrary has pointed out.
If Boca Chica were to become the production site and/or launch site for orbital vehicles then I would assume we will see the construction of larger permanent facilities. But we’ll have to wait and see. I may have seen a comment from Elon about already building orbital vehicles, if so then my theory may not be right but then again we are having to parse some fairly terse comments from Elon. I think some people have been interpreting “flight” hardware as meaning orbital, whereas I think it can just as easily mean low altitude hopper flights. Also keep in mind that Elon stated previously that full development resources won’t be shifted to Starship until after commercial crew is up and running. I think that’s one reason why the Starship program currently has an almost whimsical appearance to it. But looks are deceiving and I’m sure the engineering behind what is currently going on is quite deliberate and more sophisticated than it may appear on the surface.
Shipping such a large hull will not be easy, i suspect they will continue to be assembled at the launch site, whether it is Boca Chica or Cape Canaveral.
Musk has indicated that Boca Chic will become the primary launch site. Bad news for those of us favoring Florida.
You’re missing that much of the build is being done by Caldwell, a very experienced company which builds high end tanks, big ones, sometimes using automated seam welders like those at Boca Chica. The “pros from Dover.”
http://www.caldwelltanks.com
You don’t often see an orbital launch vehicle being hand-assembled out in the open, outdoors. Definitely a WTF, cognitive-dissonance whiplash moment.
On the other hand, people were saying the same sorts of things about vertical landing of a reusable first stage in 2010. Sometimes that intuitive whiplash is a mistaken, gut feeling. (More of the gravy than the grave, as Charles Dickens put in a very different context.)
Indeed. I wasn’t criticizing Elon’s approach…merely making a cheeky observation.
There’s a cool Robert Heinlein vibe to watching anyone build a serious spacecraft outdoors. You expect to hear dialogue like, “Gimme a three-eighths, Jimmy.” “Hey. back up the crane just a tick so this panel fits in.”
Once the outer shell is assembled it can serve as its own assembly building, maybe? The interior becomes a pretty well controllable environment. Like outfitting a ship.
It’s a little weird to see it assembled in the open, but I guess maybe that works with steel? The proposal for Sea Dragon was to make it in a shipyard.
Love this! I had a discussion a few months ago with some aerospace engineers from California. I opined that this very thing would happen. They thought I was being funny…