Author Topic: Starship Manufacturing Facility in Boca Chica  (Read 844558 times)

Offline bstrong

  • Full Member
  • ****
  • Posts: 502
  • Liked: 695
  • Likes Given: 452
Re: Starship Manufacturing Facility in Boca Chica
« Reply #340 on: 09/09/2020 07:32 pm »
I'm in the no hoist or bridge crane camp. This building was put up for one reason: they couldn't get welds of acceptable quality unless they had some walls and a roof to block the wind and rain.

Bridge cranes and hoists are not strict requirements and would probably have cost a week or two in design and construction time. A week here, a week there, and before you know it, you've missed the next Mars window.

Offline meekGee

  • Senior Member
  • *****
  • Posts: 12667
  • N. California
  • Liked: 11789
  • Likes Given: 1323
Re: Starship Manufacturing Facility in Boca Chica
« Reply #341 on: 09/09/2020 07:43 pm »
I'm in the no hoist or bridge crane camp. This building was put up for one reason: they couldn't get welds of acceptable quality unless they had some walls and a roof to block the wind and rain.

Bridge cranes and hoists are not strict requirements and would probably have cost a week or two in design and construction time. A week here, a week there, and before you know it, you've missed the next Mars window.
Sure, but how much time will fixed hoists save over the next year or two?

And how will you hoist anything in the back of the building with an external crane?

And... That statement of purpose about wind protection - that's just, like, your opinion, right?
« Last Edit: 09/09/2020 07:44 pm by meekGee »
ABCD - Always Be Counting Down

Offline rsdavis9

Re: Starship Manufacturing Facility in Boca Chica
« Reply #342 on: 09/09/2020 07:57 pm »
And if load of hoisted object is too large. They move the mostly complete SH/SS out of the way, bring in new segment, hoist new segment, rolll mostly complete back in and stack. That way you are adding to the top with lower mass segments. This assumes moving objects horizontally across floor is relatively easy.
With ELV best efficiency was the paradigm. The new paradigm is reusable, good enough, and commonality of design.
Same engines. Design once. Same vehicle. Design once. Reusable. Build once.

Offline bstrong

  • Full Member
  • ****
  • Posts: 502
  • Liked: 695
  • Likes Given: 452
Re: Starship Manufacturing Facility in Boca Chica
« Reply #343 on: 09/09/2020 08:36 pm »
I'm in the no hoist or bridge crane camp. This building was put up for one reason: they couldn't get welds of acceptable quality unless they had some walls and a roof to block the wind and rain.

Bridge cranes and hoists are not strict requirements and would probably have cost a week or two in design and construction time. A week here, a week there, and before you know it, you've missed the next Mars window.
Sure, but how much time will fixed hoists save over the next year or two?

And how will you hoist anything in the back of the building with an external crane?

And... That statement of purpose about wind protection - that's just, like, your opinion, right?

Of course, that whole post (and this one) is IMO, and I also don't disagree that a hoist is a good idea. It will cost a little bit of time and money in the short term in exchange for larger projected savings in the long run. We all make decisions like that every day, and if I were in charge, I would totally have approved the hoist.

But Elon's decision process changed last year. I think he wakes up every morning thinking about how to move every critical path task on every schedule to the left. I imagine him sitting in on every design review asking, for every element of every design, "Is this strictly necessary to unblock the next task on the schedule?" If the answer is no, he nixes it.

Most good project managers do this to an extent, but I think that Elon is going really hard core on it right now. Sometimes that manifests itself in weird ways, like omitting a hoist in a building that could obviously use one. And sometimes it results in engineers going back to the drawing board to find out-of-the box ways to get the thing they want without affecting the critical path schedule.

I bring it up because I think that a lot of little decisions at Boca make more sense if you think of Elon (or some surrogate) sitting in the corner nixing random useful things that any other reasonable manager would allow just because they delay a critical path task by a day or two. This is a fun one to watch because by your perfectly reasonable logic about saving time in the long run, a hoist makes obvious sense. But I'm proposing that by Elon's logic, the only thing that matters is getting the first SH stacked as soon as possible.

Offline cdebuhr

  • Full Member
  • ****
  • Posts: 839
  • Calgary, AB
  • Liked: 1411
  • Likes Given: 589
Re: Starship Manufacturing Facility in Boca Chica
« Reply #344 on: 09/09/2020 08:39 pm »
I'm in the no hoist or bridge crane camp. This building was put up for one reason: they couldn't get welds of acceptable quality unless they had some walls and a roof to block the wind and rain.

Bridge cranes and hoists are not strict requirements and would probably have cost a week or two in design and construction time. A week here, a week there, and before you know it, you've missed the next Mars window.
Sure, but how much time will fixed hoists save over the next year or two?

And how will you hoist anything in the back of the building with an external crane?

And... That statement of purpose about wind protection - that's just, like, your opinion, right?
Who said anything about an external crane?  As near as I can tell, each side of the HB has a footprint equal (more or less) to the MB, with lots of room right up the middle for a mobile crane/roll lift.  No reason they have to stay outside.  They do it that way with the MB because there isn't room.

Offline gemmy0I

  • Full Member
  • **
  • Posts: 268
  • Liked: 503
  • Likes Given: 1703
Re: Starship Manufacturing Facility in Boca Chica
« Reply #345 on: 09/09/2020 09:53 pm »
Of course, that whole post (and this one) is IMO, and I also don't disagree that a hoist is a good idea. It will cost a little bit of time and money in the short term in exchange for larger projected savings in the long run. We all make decisions like that every day, and if I were in charge, I would totally have approved the hoist.

But Elon's decision process changed last year. I think he wakes up every morning thinking about how to move every critical path task on every schedule to the left. I imagine him sitting in on every design review asking, for every element of every design, "Is this strictly necessary to unblock the next task on the schedule?" If the answer is no, he nixes it.

Most good project managers do this to an extent, but I think that Elon is going really hard core on it right now. Sometimes that manifests itself in weird ways, like omitting a hoist in a building that could obviously use one. And sometimes it results in engineers going back to the drawing board to find out-of-the box ways to get the thing they want without affecting the critical path schedule.

I bring it up because I think that a lot of little decisions at Boca make more sense if you think of Elon (or some surrogate) sitting in the corner nixing random useful things that any other reasonable manager would allow just because they delay a critical path task by a day or two. This is a fun one to watch because by your perfectly reasonable logic about saving time in the long run, a hoist makes obvious sense. But I'm proposing that by Elon's logic, the only thing that matters is getting the first SH stacked as soon as possible.
This is an excellent point - Elon's philosophy of extreme rapid iteration definitely leads to some "out-of-the-box" thinking on long-term vs. short-term design choices (based on the principle that getting tangible experience sooner will help drive a better long-term design).

That said, he seems to swing back and forth frequently on exactly how to balance this philosophy with his big-picture goals, which are nothing if not long-term. This is more art than science and he (and the company as a whole) can only learn it through experience, especially since they are intentionally rejecting a lot of the conventional "rules of thumb" from an aerospace industry that's swung way too far in the "get it right the first time" direction. (There's an interesting interplay here because they're trying to revolutionize an industry that, paradoxically, thinks too long-term with its tactical design choices, e.g. production line design, whilst thinking too short-term in its strategic ambitions, e.g. incremental designs like Atlas VI Vulcan instead of something world-changing like Starship.)

They got burned somewhat by thinking too short-term with the Mk1 prototype build. By focusing so much on getting that first prototype to "completion" ASAP, they ended up with a bespoke, low-fidelity prototype that clearly taught them far less about the eventual flight design than they were hoping it would. Since then, Elon's emphasized over and over that their focus is now on building an assembly line to mass-produce Starships, rather than particular Starship prototypes; the prototypes are a byproduct of that (albeit a critical one that feeds back into the overall process).

That seems to represent a degree of push-back against a pure "critical path at all costs" mindset. All that excruciating time "wasted" manually polishing Mk1's welds on cherry pickers was time and effort they could have spent building an assembly line to produce many more prototypes. (There was a rumor at the time, claiming to be from an insider at Boca Chica, that Elon "blew a gasket" shortly after the big presentation last year when he learned that, after everything they'd sunk into Mk1, it was unlikely to survive a 20km hop and had done next to nothing to produce a true assembly line. Shortly after that was when he basically moved into a house in Boca Chica Village and took much more personal oversight of the site, something he previously hadn't been able to do while being focused on Model 3 production at Tesla.)

In other words, they are indeed keeping a laser-focus on the critical path, but it's the critical path for Mars, not for individual prototypes or testing milestones. The lesson they took away from Mk1 is clearly that being able to mass-produce Starships as efficiently as possible is essential to the big-picture critical path. That lesson has been driven home several times since, as they've found themselves waiting on new, "unexpected" critical path blockers thanks to prototypes underperforming or being lost to unexpected/"silly" mishaps. (e.g. having to wait on rebuilding the launch stand after SN4 blew up, with basically-complete SN5 and SN6 held up waiting on that) Being able to rapidly crank out a replacement for a failed (or usefully expended) test article saves them far more on the overall critical path than going all-in on a particular step in the process.

Elon seems to have gone back and forth quite a bit on this philosophy over the years as he's experimented to find the optimal "happy medium". The Tesla Model 3's extended period of "production hell" while they were trying to work out the bugs in a forward-thinking but over-automated production line clearly weighed heavily on him, and gave us the radically un-developed production system for Mk1. When that proved to be too narrow-focused, he stepped back and started building a systematic assembly line for Starships - albeit making heavy use of temporary structures and "un-processes" as proverbial scaffolding to allow them to learn as they went, instead of investing too early in processes that would quickly become outdated.

So far, they seem to be finding a good happy medium, gradually making the Starship line more systematic while continuing to crank out useful prototypes at a rapid enough pace to allow feedback into the development process. But it's clear they're still learning lessons and frequently kicking themselves when they discover weaknesses in the process - for instance, losing the launch stand in the SN4 conflagration and having testing grind to a halt while they rebuilt it. They've clearly put a lot of thought into making the replacement launch stand more robust, and the fact that they seem to be building a second one in advance of needing it points to a healthy fear of getting stuck again if they lose one.

All of which is to say, I have no idea whether they'll go with a hoist, a bridge crane, or whatever in the new High Bay. :) But I wouldn't count out the possibility of them thinking more long-term at this stage - not as an admission of failure in their simplified construction philosophy as some have suggested, but as the intended result of that process. We're seeing an increasingly systematic, well-developed assembly line take shape. They're increasingly choosing semi-permanent steel frame buildings instead of "tents", even for smaller structures like the ground fabrication building that are well within the capabilities of Sprung structures. If a crane/hoist/etc. in the High Bay will help them substantially increase their production rate of Starships and/or Super Heavies in the medium term, they'll quite likely deem it worthwhile even if it delays the first Super Heavy stacking by a few weeks - because they'll lose far more time if (when!) those early prototypes go boom if they don't have more immediately waiting in the wings to pick up the baton.

Offline Dave G

  • Senior Member
  • *****
  • Posts: 3231
  • Liked: 2125
  • Likes Given: 2021
Re: Starship Manufacturing Facility in Boca Chica
« Reply #346 on: 09/09/2020 10:20 pm »
But Elon's decision process changed last year. I think he wakes up every morning thinking about how to move every critical path task on every schedule to the left. I imagine him sitting in on every design review asking, for every element of every design, "Is this strictly necessary to unblock the next task on the schedule?" If the answer is no, he nixes it.

My opinion: The rapid iterations we're seeing aren't about shortening the schedule.
They're about learning how to build it.
As Robert Zubrin said, Musk is not really building a ship, he's building a shipyard.


For me, the following quote is the key to understanding Musk's development strategy:
Quote from: Elon Musk 2020-02-28
When you design the object at first you don't realize all the parts that are difficult to manufacture.
So having the manufacturing system and the design, bring those up at the same time,
so that you're actually at the beginning making a thing that you know is wrong.
But you're actually figuring out what's hard to manufacture. That's the real problem...
And then we discover, oh there's all these things in the design that are very difficult to make, and so therefore we must change the design.

Musk wants to colonize Mars. That's his goal.
To do that, he needs hundreds or thousands of Starships.
For this, Starship's design needs to be relatively easy to manufacture.

So I think that's what's driving the rapid iterations we're seeing.
Musk wants to crank out a least one Starship a week, preferably two a week.
And he want's to get to that point as quickly as possible.

If he was concerned with getting the first Starship to orbit ASAP, I don't think we would see so many iterations.

[Edit] Fixed first quote author.
« Last Edit: 09/10/2020 10:34 am by Dave G »

Offline gemmy0I

  • Full Member
  • **
  • Posts: 268
  • Liked: 503
  • Likes Given: 1703
Re: Starship Manufacturing Facility in Boca Chica
« Reply #347 on: 09/09/2020 11:49 pm »
But Elon's decision process changed last year. I think he wakes up every morning thinking about how to move every critical path task on every schedule to the left. I imagine him sitting in on every design review asking, for every element of every design, "Is this strictly necessary to unblock the next task on the schedule?" If the answer is no, he nixes it.

My opinion: The rapid iterations we're seeing aren't about shortening the schedule.
They're about learning how to build it.
As Robert Zubrin said, Musk is not really building a ship, he's building a shipyard.


For me, the following quote is the key to understanding Musk's development strategy:
Quote from: Elon Musk 2020-02-28
When you design the object at first you don't realize all the parts that are difficult to manufacture.
So having the manufacturing system and the design, bring those up at the same time,
so that you're actually at the beginning making a thing that you know is wrong.
But you're actually figuring out what's hard to manufacture. That's the real problem...
And then we discover, oh there's all these things in the design that are very difficult to make, and so therefore we must change the design.

Musk wants to colonize Mars. That's his goal.
To do that, he needs hundreds or thousands of Starships.
For this, Starship's design needs to be relatively easy to manufacture.

So I think that's what's driving the rapid iterations we're seeing.
Musk wants to crank out a least one Starship a week, preferably two a week.
And he want's to get to that point as quickly as possible.

If he was concerned with getting the first Starship to orbit ASAP, I don't think we would see so many iterations.
(FYI, you might want to fix your quotes (the first paragraph you quoted was from bstrong, not me - guessing you quoted it from my response). ;) )

Excellent point - although I would submit that learning how to build the shipyard, the ship, and shortening the schedule all go hand-in-hand. I think what we're seeing, with all its iterations, actually is the quickest way to get the first Starship to orbit - if by "Starship" we mean a viable commercial product that can actually deliver on the goal of rapid reusability, rather than a risky Mk1-style prototype that no customer would ever fly on and would require months to refurbish (if it could even stick a landing). That's not just important for the goal of sending thousands of Starships to Mars; that's vital even for less ambitious goals like supporting Artemis, the Dear Moon project, or even just fulfilling their FCC obligations to get Starlink up.

Gradually transitioning from a loose "un-process" that minimizes cost sinkage into infrastructure "that you know is wrong", to an increasingly well-developed assembly line capable of producing enough test articles to maintain the process of discovering "what's wrong", is the quickest way to get to that end goal. It honestly acknowledges that any endeavor of great scope (which even the most basic LEO version of Starship is in and of itself, even discounting their Mars ambitions) is fraught with unknowns both known and unknown, and that giants must be raised so that greater giants may stand on their shoulders. Cutting corners in that process in the hopes of saving time and money falls into the "traditional aerospace" hubris of thinking one can "get it right the first time". When things do inevitably turn out to be "wrong", fragile timelines and budgets are broken by having to respond to that reality without being prepared for it (e.g., polishing Mk1's sloppy welds, rebuilding the launch stand after SN4, or that time SLS had to set aside an entire set of tanks due to welding issues) - made worse by the human tendency to stick with a broken plan rather than admit it was flawed and acknowledge sunk costs as losses.

These are general principles of design and engineering that apply across all fields, anywhere the inadequacy of human knowledge comes into play. Even in the pure sciences, imperfect theories are needed to help us learn what's lacking in them. And in the creative arts, writers, painters, etc. go through many drafts to build their way to a final product. Heck, even writing this post required a few passes. ;) Other fields besides aerospace have had to painfully undergo (and are in some ways still undergoing) revolutions to realize this, e.g. software engineering's switch from the "waterfall model" to agile development over the last several decades, which itself was inspired by continuous improvement and lean manufacturing philosophies developed in physical engineering by companies like Toyota.

Somehow, fields that consider themselves "mission-critical" managed to escape that revolution through the thinking that the sensitivity of their designs required fully pre-planned processes. This is all too common not just in aerospace but in anything that's considered "mission-critical", whether it be fighter jet avionics, a nuclear power plant, or a national online healthcare marketplace. The sad reality is that processes that aren't prepared to rapidly iterate will not only end up over budget and behind schedule, but will produce substantially lower-quality products. SpaceX and its "new space" brethren are finally learning what that revolution needs to look like for aerospace, and we're getting a wonderful view of that discovery process (with all its missteps and "learning moments") streamed live by webcam from South Texas. :)

There's no escaping the fact that we just can't know the full extent of what we don't yet know. A project that acknowledges that truth will, as a general rule, be able to get to its goal faster than one that tries to plow ahead based on a preliminary conception of the task. Less iteration might've gotten something approximating Starship to orbit a little sooner, but the price would've been paid down the road in refurbishment hell and painful "down periods" where they discovered all the ways they "got lucky" the first time and had to re-work major aspects of the design - all on the ultimate critical path of the vehicle's operational service. That's what happened with Shuttle, with Angara (which, shockingly, is still stuck in development hell despite successfully flying in 2014), with Starliner, and based on history, will probably happen with SLS.

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 4327
  • Terra is my nation; currently Kansas
  • Liked: 2951
  • Likes Given: 4855
Re: Starship Manufacturing Facility in Boca Chica
« Reply #348 on: 09/10/2020 02:15 am »
I'm in the no hoist or bridge crane camp. This building was put up for one reason: they couldn't get welds of acceptable quality unless they had some walls and a roof to block the wind and rain.

Bridge cranes and hoists are not strict requirements and would probably have cost a week or two in design and construction time. A week here, a week there, and before you know it, you've missed the next Mars window.
Using external cranes seems to be a 'git er dun' solution while you're figuring how you want to dun it. A bridge crane might add a week or two but a simple hoist is something hobbyists install to pull an engine.


How much time do they loose in the mid bay on that last lift for the skirt bottom? It's a touchy squeeze. How much is crane rental? How much subtle traffic interference is there with that bigazzed crane camped out front?


A hoist is so cheap, fast to install and easy to use it makes too much sense.
We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Offline stcks

  • Full Member
  • **
  • Posts: 252
  • Liked: 266
  • Likes Given: 312
Re: Starship Manufacturing Facility in Boca Chica
« Reply #349 on: 09/10/2020 02:21 am »
a 'git er dun' solution while you're figuring how you want to dun it.

😂 I'm stealing this

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 4327
  • Terra is my nation; currently Kansas
  • Liked: 2951
  • Likes Given: 4855
Re: Starship Manufacturing Facility in Boca Chica
« Reply #350 on: 09/10/2020 02:29 am »
I'm in the no hoist or bridge crane camp. This building was put up for one reason: they couldn't get welds of acceptable quality unless they had some walls and a roof to block the wind and rain.

Bridge cranes and hoists are not strict requirements and would probably have cost a week or two in design and construction time. A week here, a week there, and before you know it, you've missed the next Mars window.
Sure, but how much time will fixed hoists save over the next year or two?

And how will you hoist anything in the back of the building with an external crane?

And... That statement of purpose about wind protection - that's just, like, your opinion, right?

Of course, that whole post (and this one) is IMO, and I also don't disagree that a hoist is a good idea. It will cost a little bit of time and money in the short term in exchange for larger projected savings in the long run. We all make decisions like that every day, and if I were in charge, I would totally have approved the hoist.

But Elon's decision process changed last year. I think he wakes up every morning thinking about how to move every critical path task on every schedule to the left. I imagine him sitting in on every design review asking, for every element of every design, "Is this strictly necessary to unblock the next task on the schedule?" If the answer is no, he nixes it.

Most good project managers do this to an extent, but I think that Elon is going really hard core on it right now. Sometimes that manifests itself in weird ways, like omitting a hoist in a building that could obviously use one. And sometimes it results in engineers going back to the drawing board to find out-of-the box ways to get the thing they want without affecting the critical path schedule.

I bring it up because I think that a lot of little decisions at Boca make more sense if you think of Elon (or some surrogate) sitting in the corner nixing random useful things that any other reasonable manager would allow just because they delay a critical path task by a day or two. This is a fun one to watch because by your perfectly reasonable logic about saving time in the long run, a hoist makes obvious sense. But I'm proposing that by Elon's logic, the only thing that matters is getting the first SH stacked as soon as possible.
The high bay is ticking right along. It could have started right after the mid bay but it didn't because it wasn't needed then. High bay will probably be ready to stack an SH before the launch stand is ready. And that's only because tying in the rebar and sleeving it turned out to be harder than expected. Otherwise it would be near a dead heat. JIT.


Now that they know what they need to do I don't think they need to be so tight with time that they are tripping over themselves.
We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Offline Nomadd

  • Senior Member
  • *****
  • Posts: 8436
  • Highway Whatever
  • Liked: 57163
  • Likes Given: 1103
Re: Starship Manufacturing Facility in Boca Chica
« Reply #351 on: 09/10/2020 03:26 am »
a 'git er dun' solution while you're figuring how you want to dun it.

😂 I'm stealing this
Those who danced were thought to be quite insane by those who couldn't hear the music.

Offline OTV Booster

  • Senior Member
  • *****
  • Posts: 4327
  • Terra is my nation; currently Kansas
  • Liked: 2951
  • Likes Given: 4855
Re: Starship Manufacturing Facility in Boca Chica
« Reply #352 on: 09/10/2020 02:18 pm »
Of course, that whole post (and this one) is IMO, and I also don't disagree that a hoist is a good idea. It will cost a little bit of time and money in the short term in exchange for larger projected savings in the long run. We all make decisions like that every day, and if I were in charge, I would totally have approved the hoist.

But Elon's decision process changed last year. I think he wakes up every morning thinking about how to move every critical path task on every schedule to the left. I imagine him sitting in on every design review asking, for every element of every design, "Is this strictly necessary to unblock the next task on the schedule?" If the answer is no, he nixes it.

Most good project managers do this to an extent, but I think that Elon is going really hard core on it right now. Sometimes that manifests itself in weird ways, like omitting a hoist in a building that could obviously use one. And sometimes it results in engineers going back to the drawing board to find out-of-the box ways to get the thing they want without affecting the critical path schedule.

I bring it up because I think that a lot of little decisions at Boca make more sense if you think of Elon (or some surrogate) sitting in the corner nixing random useful things that any other reasonable manager would allow just because they delay a critical path task by a day or two. This is a fun one to watch because by your perfectly reasonable logic about saving time in the long run, a hoist makes obvious sense. But I'm proposing that by Elon's logic, the only thing that matters is getting the first SH stacked as soon as possible.
This is an excellent point - Elon's philosophy of extreme rapid iteration definitely leads to some "out-of-the-box" thinking on long-term vs. short-term design choices (based on the principle that getting tangible experience sooner will help drive a better long-term design).

That said, he seems to swing back and forth frequently on exactly how to balance this philosophy with his big-picture goals, which are nothing if not long-term. This is more art than science and he (and the company as a whole) can only learn it through experience, especially since they are intentionally rejecting a lot of the conventional "rules of thumb" from an aerospace industry that's swung way too far in the "get it right the first time" direction. (There's an interesting interplay here because they're trying to revolutionize an industry that, paradoxically, thinks too long-term with its tactical design choices, e.g. production line design, whilst thinking too short-term in its strategic ambitions, e.g. incremental designs like Atlas VI Vulcan instead of something world-changing like Starship.)

They got burned somewhat by thinking too short-term with the Mk1 prototype build. By focusing so much on getting that first prototype to "completion" ASAP, they ended up with a bespoke, low-fidelity prototype that clearly taught them far less about the eventual flight design than they were hoping it would. Since then, Elon's emphasized over and over that their focus is now on building an assembly line to mass-produce Starships, rather than particular Starship prototypes; the prototypes are a byproduct of that (albeit a critical one that feeds back into the overall process).

That seems to represent a degree of push-back against a pure "critical path at all costs" mindset. All that excruciating time "wasted" manually polishing Mk1's welds on cherry pickers was time and effort they could have spent building an assembly line to produce many more prototypes. (There was a rumor at the time, claiming to be from an insider at Boca Chica, that Elon "blew a gasket" shortly after the big presentation last year when he learned that, after everything they'd sunk into Mk1, it was unlikely to survive a 20km hop and had done next to nothing to produce a true assembly line. Shortly after that was when he basically moved into a house in Boca Chica Village and took much more personal oversight of the site, something he previously hadn't been able to do while being focused on Model 3 production at Tesla.)

In other words, they are indeed keeping a laser-focus on the critical path, but it's the critical path for Mars, not for individual prototypes or testing milestones. The lesson they took away from Mk1 is clearly that being able to mass-produce Starships as efficiently as possible is essential to the big-picture critical path. That lesson has been driven home several times since, as they've found themselves waiting on new, "unexpected" critical path blockers thanks to prototypes underperforming or being lost to unexpected/"silly" mishaps. (e.g. having to wait on rebuilding the launch stand after SN4 blew up, with basically-complete SN5 and SN6 held up waiting on that) Being able to rapidly crank out a replacement for a failed (or usefully expended) test article saves them far more on the overall critical path than going all-in on a particular step in the process.

Elon seems to have gone back and forth quite a bit on this philosophy over the years as he's experimented to find the optimal "happy medium". The Tesla Model 3's extended period of "production hell" while they were trying to work out the bugs in a forward-thinking but over-automated production line clearly weighed heavily on him, and gave us the radically un-developed production system for Mk1. When that proved to be too narrow-focused, he stepped back and started building a systematic assembly line for Starships - albeit making heavy use of temporary structures and "un-processes" as proverbial scaffolding to allow them to learn as they went, instead of investing too early in processes that would quickly become outdated.

So far, they seem to be finding a good happy medium, gradually making the Starship line more systematic while continuing to crank out useful prototypes at a rapid enough pace to allow feedback into the development process. But it's clear they're still learning lessons and frequently kicking themselves when they discover weaknesses in the process - for instance, losing the launch stand in the SN4 conflagration and having testing grind to a halt while they rebuilt it. They've clearly put a lot of thought into making the replacement launch stand more robust, and the fact that they seem to be building a second one in advance of needing it points to a healthy fear of getting stuck again if they lose one.

All of which is to say, I have no idea whether they'll go with a hoist, a bridge crane, or whatever in the new High Bay. :) But I wouldn't count out the possibility of them thinking more long-term at this stage - not as an admission of failure in their simplified construction philosophy as some have suggested, but as the intended result of that process. We're seeing an increasingly systematic, well-developed assembly line take shape. They're increasingly choosing semi-permanent steel frame buildings instead of "tents", even for smaller structures like the ground fabrication building that are well within the capabilities of Sprung structures. If a crane/hoist/etc. in the High Bay will help them substantially increase their production rate of Starships and/or Super Heavies in the medium term, they'll quite likely deem it worthwhile even if it delays the first Super Heavy stacking by a few weeks - because they'll lose far more time if (when!) those early prototypes go boom if they don't have more immediately waiting in the wings to pick up the baton.
Bravo! Well said. You have drawn several threads together and shown them to be related parts of the whole. I believe you have nailed it.
We are on the cusp of revolutionary access to space. One hallmark of a revolution is that there is a disjuncture through which projections do not work. The thread must be picked up anew and the tapestry of history woven with a fresh pattern.

Offline Nonproliferator

  • Member
  • Posts: 35
  • Liked: 54
  • Likes Given: 41
Re: Starship Manufacturing Facility in Boca Chica
« Reply #353 on: 09/10/2020 03:28 pm »
This discussion has been illuminating but I have a quick question related to this matter of internal vs external craning for the high bay.

As illustrated by the mid-bay there is a need for considerable head-room in order to allow an external crane to lift the elements of the starship builds, and things get pretty tight at the final stage. All of the visualization of the SH in the highbay seen to date suggest there really isn't a lot of space between the top of a SH on a stand, and the ceiling of the high-bay.

So, there is an actual question in here, is there actually sufficient clearance for the use of an external crane in the highbay to build SH anyway? At least at the final step.

As an aside, the "obvious" solution in the absence of sufficient headroom would seem to be something like allowing the central section of the roof to be opened at need.

Offline wannamoonbase

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 4799
  • Denver, CO
    • U.S. Metric Association
  • Liked: 2532
  • Likes Given: 3233
Re: Starship Manufacturing Facility in Boca Chica
« Reply #354 on: 09/10/2020 05:23 pm »
All of which is to say, I have no idea whether they'll go with a hoist, a bridge crane, or whatever in the new High Bay. :) But I wouldn't count out the possibility of them thinking more long-term at this stage - not as an admission of failure in their simplified construction philosophy as some have suggested, but as the intended result of that process. We're seeing an increasingly systematic, well-developed assembly line take shape. They're increasingly choosing semi-permanent steel frame buildings instead of "tents", even for smaller structures like the ground fabrication building that are well within the capabilities of Sprung structures. If a crane/hoist/etc. in the High Bay will help them substantially increase their production rate of Starships and/or Super Heavies in the medium term, they'll quite likely deem it worthwhile even if it delays the first Super Heavy stacking by a few weeks - because they'll lose far more time if (when!) those early prototypes go boom if they don't have more immediately waiting in the wings to pick up the baton.

I think your original post is really good.

I think we can look at the hodge podge of buildings at BC as the 0.9 or 1.0 assembly facility.  Keep it cheap and flexible, add what you need when you need it.

Elon looks at Tesla Gigafactorys as a product in themselves.  SS/SH assembly is likely the same idea in his head.

Once they are mature and sorted I expect we'd see a refined more capable building(s).  1 large building with side bays that make heads and rings, those parts feed into mid bays, larger assemblies move into high pays for assembly and detailing.

All under 1 roof with bridge cranes and assembly equipment everywhere to make moving things around as easy as possible.  With automation where possible.

It's not that much of a stretch to think about Elon imagining a rocket factory that poops out a SS or SH once per shift.  (Not saying that will happen or is needed.  But he could easily be thinking about it.)
« Last Edit: 09/10/2020 05:29 pm by wannamoonbase »
Superheavy + Starship the final push to launch commit!

Offline testguy

  • Full Member
  • ****
  • Posts: 595
  • Clifton, Virginia
  • Liked: 617
  • Likes Given: 567
Re: Starship Manufacturing Facility in Boca Chica
« Reply #355 on: 09/12/2020 12:09 am »
Just watched Mary’s latest video.  Just wanted to tip my hat to those construction workers working on the high bay.  They were in a bucket high up being blown in the wind a lot.  Those guys are much braver than me.











Offline enbandi

  • Full Member
  • ****
  • Posts: 652
  • Hungary
  • Liked: 786
  • Likes Given: 385
Re: Starship Manufacturing Facility in Boca Chica
« Reply #356 on: 09/12/2020 05:29 pm »
The critical-path and not really necessary arguments are really good for the lack of internal hoists/cranes for the Mid Bay. Not in terms of the days or a week needed to istall something like that, but when the Mid Bay started there were rumors about that is a repurposed building, originally designed and manufactured for something (different) at Florida.

That speculation may be too wild, but if they pushed that building real fast, and had to choose between a tailored (heavy construction) with a crane 2-3 months later, or the current Mid Bay using prefab pieces, or existing plans (without a crane) that could be a sensible explanation why they use external cranes there.

But in my opinion the same argument does not hold for the High Bay. Its a much longer lead time item, obviously designed and manufactured for this exact usage at BC. So either with or without crane, the explanation shall be different.

Offline meekGee

  • Senior Member
  • *****
  • Posts: 12667
  • N. California
  • Liked: 11789
  • Likes Given: 1323
Re: Starship Manufacturing Facility in Boca Chica
« Reply #357 on: 09/12/2020 06:32 pm »
Hmmm looking at the pictures, I still want 4 stationary hoists, but they're leaving room for potentially a short-travel bridge crane under the central area of the roof...
« Last Edit: 09/12/2020 06:33 pm by meekGee »
ABCD - Always Be Counting Down

Offline mmeijeri

  • Senior Member
  • *****
  • Posts: 7767
  • Martijn Meijering
  • NL
  • Liked: 387
  • Likes Given: 795
Re: Starship Manufacturing Facility in Boca Chica
« Reply #358 on: 09/12/2020 06:33 pm »
Would two forward-backward travelling cranes be an option?
« Last Edit: 09/12/2020 06:34 pm by mmeijeri »
Pro-tip: you don't have to be a jerk if someone doesn't agree with your theories

Offline enbandi

  • Full Member
  • ****
  • Posts: 652
  • Hungary
  • Liked: 786
  • Likes Given: 385
Re: Starship Manufacturing Facility in Boca Chica
« Reply #359 on: 09/12/2020 06:48 pm »
I wonder what is that bracket for? (High Bay, middle apron, back panel, just under the roof level).

Best candidate for anything crane related so faar.

Orig image from Nomadd.

Edit: orig images in this post
Getting some pretty sturdy looking roof beams over the middle of Neo Highbay.
« Last Edit: 09/12/2020 06:52 pm by enbandi »

 

Advertisement NovaTech
Advertisement SkyTale Software GmbH
Advertisement Northrop Grumman
Advertisement
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0