Starship is Go...

Oldfart1939 · 2020-05-12 21:29:46

Increasing complexity of the product tends to diminish the savings of "mass production." Simply put, increase in numbers of a complex product, as GW has pointed out, smooths the quality a lot and ups need for rigorous quality control. Cost savings are only moderate.
Making repetitive numbers of an object--call it a product--allows the workforce to become highly refined in it's skill set. Note improved appearance of each successive Starship. Welders are learning how to weld these particular structures and practice makes perfect.

louis · 2020-05-13 06:06:15

You can't get much more complicated than a commercial airliner - they have over a million parts. But Boeing still find it saves time and therefore money to put them on a production line.

I would think that one advantage of a production line for Space X will be that they can use industrial robots for the welding process.

Oldfart1939 wrote:

Increasing complexity of the product tends to diminish the savings of "mass production." Simply put, increase in numbers of a complex product, as GW has pointed out, smooths the quality a lot and ups need for rigorous quality control. Cost savings are only moderate.
Making repetitive numbers of an object--call it a product--allows the workforce to become highly refined in it's skill set. Note improved appearance of each successive Starship. Welders are learning how to weld these particular structures and practice makes perfect.

SpaceNut · 2020-05-13 07:12:55

Low production rates also cause a retaining of employees at a higher cost since they have the skills required.

kbd512 · 2020-05-13 11:15:53

Louis,

Robotic welding equipment might save some time on fabrication and therefore cost of the propellant tanks, but the engines / avionics / software / heat shield, as always, are the most complex parts on the vehicle. I said "might" because even though the welding time is reduced by using good fixtures and jigs, laying up the parts within the tolerances required in those jigs is also a time consuming process. QC is very important at that point, as GW pointed out, else the parts produced will be unusable- there's no "weld-check-weld" at that point, you either get it right from the word "go" or you scrap the entire part in many cases. The benefit is that welding robots are not as expensive, long term, as a standing army of competent aerospace welders. Mass production of the propellant tanks becomes feasible, but at significant cost. That's why the production run has to be significant, in order to justify the costs associated with that level of process automation. That said, rockets and aircraft are inherently touch-labor intensive and a largely automated factory like GM's or BMW's automotive factories would be cost-prohibitive.

louis · 2020-05-13 11:45:44

I agree it has to be a significant production run and maybe one a week will put Space X in that ballpark. Another advantage of a production line for this sort of production is that you don't have to move manufacturing machinery into and out of the zone all the time - there's far less dead time of that nature.

kbd512 wrote:

Louis,
Robotic welding equipment might save some time on fabrication and therefore cost of the propellant tanks, but the engines / avionics / software / heat shield, as always, are the most complex parts on the vehicle. I said "might" because even though the welding time is reduced by using good fixtures and jigs, laying up the parts within the tolerances required in those jigs is also a time consuming process. QC is very important at that point, as GW pointed out, else the parts produced will be unusable- there's no "weld-check-weld" at that point, you either get it right from the word "go" or you scrap the entire part in many cases. The benefit is that welding robots are not as expensive, long term, as a standing army of competent aerospace welders. Mass production of the propellant tanks becomes feasible, but at significant cost. That's why the production run has to be significant, in order to justify the costs associated with that level of process automation. That said, rockets and aircraft are inherently touch-labor intensive and a largely automated factory like GM's or BMW's automotive factories would be cost-prohibitive.

louis · 2020-05-15 11:41:51

Latest from Felix...

https://www.youtube.com/watch?v=-k5m7U3CbRo&t=238s

At 6:05 - seems like they have already started some robot welding at Boca Chica (shape of things to come, I think).

louis · 2020-05-19 16:43:13

https://www.youtube.com/watch?v=kICMCCXAKVg

Brilliant! SN4 survives Static Fire No. 2...bit of after-burn though!

GW Johnson · 2020-05-21 11:41:59

Venting post-fire is normal for cryogenics. In this case, one of those is essentially natural gas. One should not be surprised to see it ignite into a flare, much like the flare adjacent to the test stand in the video. As long as gas is being vented, you may or may not be successful extinguishing the flare-off with water afterward, and they were relatively unsuccessful doing exactly that. It becomes a problem only if something is overheating because of the flare. The real end to a flare is shutting off the gas flow. But with a propellant tank containing unused liquid methane, that is not possible. Not until you empty the tank.

GW

louis · 2020-05-21 16:50:42

https://www.youtube.com/watch?v=BEWCTrWGb54

Felix looks at the issue. Does seem fairly serious. They might be having problems depressurising.

GW Johnson wrote:

Venting post-fire is normal for cryogenics. In this case, one of those is essentially natural gas. One should not be surprised to see it ignite into a flare, much like the flare adjacent to the test stand in the video. As long as gas is being vented, you may or may not be successful extinguishing the flare-off with water afterward, and they were relatively unsuccessful doing exactly that. It becomes a problem only if something is overheating because of the flare. The real end to a flare is shutting off the gas flow. But with a propellant tank containing unused liquid methane, that is not possible. Not until you empty the tank.
GW

elderflower · 2020-05-23 13:17:56

You have to flare it steadily. Abrupt venting of large tanks requires a flare system with condensate separators and tanks, surge capacity, igniters, pilot burners, inert purge and coolant sprays. Without that you get liquid slugs or fog resulting in loss of flame followed by a blevy.

GW Johnson · 2020-05-25 14:54:08

I just re-did my reverse-engineering analysis of the Starship/Superheavy vehicle. It's posted today (5-25-20) over at "exrocketman, under the title 2020 Reverse Engineering Estimates for the Starship/Superheavy.

Spacex is just claiming 100+ metric tons to LEO. I am showing that number might be 160 tons. The weakest links in my analysis are the inert masses for Starship and for Superheavy. For Starship, I used the 120 tons Musk claimed for the prototype in his Boca Chica talk, even though he said the goal was 100 tons. He also said the 85 tons shown on the slide was just wrong.

For Superheavy, I guessed a nominal 250 ton inert mass that is just about 5% of a gross liftoff mass that is just over 5000 tons. That's a 5% inert fraction for the stage, not very different from the Falcon stages.

Just for reference, NASA's Users' Guide for SLS shows LEO payload as 70 metric tons Block 1, 105 metric tons Block 1B, and 130 metric tons Block 2B. Bridenstine says he thinks a Block 1 launch will cost about $1.5B.

I took a wild guess for a Starship/Superheavy launch pricing out at about twice a Falcon-Heavy launch: about $200M. Even if I underestimated by a factor of 2, that's still a nice low price. SS/SH delivers more to LEO than any version of SLS, and prices-out 5-10 times cheaper per ton to LEO than SLS.

Man, I really do hope they succeed, and soon! What a deal that would be!

GW

Last edited by GW Johnson (2020-05-25 14:55:31)

louis · 2020-05-25 19:39:02

I was reading how they had to play powerful water hoses over the Saturn V launch pad. That wasn't to stop a fire but to stop the Saturn V rocket vibrations destroying the launch pad.

You'd think the same would be required of the Starship and in Starship's case, for a static fire, it might be safer!

elderflower wrote:

You have to flare it steadily. Abrupt venting of large tanks requires a flare system with condensate separators and tanks, surge capacity, igniters, pilot burners, inert purge and coolant sprays. Without that you get liquid slugs or fog resulting in loss of flame followed by a blevy.

Last edited by louis (2020-05-25 20:06:10)

louis · 2020-05-25 19:40:50

160 tons makes everything easier...maybe that was why their Chief Engineer was a bit cagey about how many Starship launches will be required to refuel one Starship in LEO for the Mars mission.

GW Johnson wrote:

I just re-did my reverse-engineering analysis of the Starship/Superheavy vehicle. It's posted today (5-25-20) over at "exrocketman, under the title 2020 Reverse Engineering Estimates for the Starship/Superheavy.
Spacex is just claiming 100+ metric tons to LEO. I am showing that number might be 160 tons. The weakest links in my analysis are the inert masses for Starship and for Superheavy. For Starship, I used the 120 tons Musk claimed for the prototype in his Boca Chica talk, even though he said the goal was 100 tons. He also said the 85 tons shown on the slide was just wrong.
For Superheavy, I guessed a nominal 250 ton inert mass that is just about 5% of a gross liftoff mass that is just over 5000 tons. That's a 5% inert fraction for the stage, not very different from the Falcon stages.
Just for reference, NASA's Users' Guide for SLS shows LEO payload as 70 metric tons Block 1, 105 metric tons Block 1B, and 130 metric tons Block 2B. Bridenstine says he thinks a Block 1 launch will cost about $1.5B.
I took a wild guess for a Starship/Superheavy launch pricing out at about twice a Falcon-Heavy launch: about $200M. Even if I underestimated by a factor of 2, that's still a nice low price. SS/SH delivers more to LEO than any version of SLS, and prices-out 5-10 times cheaper per ton to LEO than SLS.
Man, I really do hope they succeed, and soon! What a deal that would be!
GW

GW Johnson · 2020-05-26 11:35:50

Well, if (1) my 160 ton figure is correct, and (2) that figure represents the propellant delivered for transfer on-orbit from a tanker (big if, that), and (3) my landing allowance estimate of 9 tons is correct, then 1191 tons to fill back up divided by 160 tons deliverable equals 7.44, meaning it would take some 7 or 8 tanker flights to fully refill the thing on orbit.

I don't think that's right. I think the tanker version is different enough from the passenger/cargo version that it's really a separate vehicle design. I don't think we know anything valid about the tankers yet. I also don't think Spacex itself knows very much about them yet, they being nothing but a paper concept design.

They won't be any more than that, until Starship-as-it-is-becoming reaches orbit the first time. And tankers are not actually needed until Starship is ready to leave orbit to go elsewhere. THAT event is farther away into the future than any of us wants it to be. Myself included.

Why do I say that?

Because the tail-to-tail refill-with-cryogenics scenario has to be experimentally demonstrated and verified on-orbit, more than a few times, before it can be routinely used. And that has to be done with a family of spaceship designs that still has yet to fly at all.

I think Starships to Mars in 2022 or 2024 is rank nonsense. There is too much to do to make this thing operational for such missions. 2026 is probably nonsense, too. But unless Musk gets slowed by bureaucrats or test catastrophes, 2028 might be a decent high-odds-against bet. I think bureaucrats and test catastrophes are actually rather likely, so the 2030's are a better, more sure bet.

GW

Last edited by GW Johnson (2020-05-26 11:42:59)

SpaceNut · 2020-05-26 16:42:51

Of course long before mars they will most likely test it for a trip to the moon to give it some shaking out mission in addition to the full application of refueling on orbit. It will also give a means to test out the spin or tumble for AG creation though we will not be testing the effects on a long term mission until we actually do one.

louis · 2020-05-26 17:30:02

I am convinced there will be no AG on Mission One to Mars. Musk wants to keep this simple.

SpaceNut wrote:

Of course long before mars they will most likely test it for a trip to the moon to give it some shaking out mission in addition to the full application of refueling on orbit. It will also give a means to test out the spin or tumble for AG creation though we will not be testing the effects on a long term mission until we actually do one.

louis · 2020-05-29 18:51:03

Sheet - that was one hell of an explosion.

https://www.youtube.com/watch?v=7RPyDPpmDAk

Not sure the authorities will be too happy about that one.

kbd512 · 2020-05-29 20:00:49

It's better that this kind of thing happens now rather than later, but the more destroyed test articles I see, the more I wonder about whether or not they need to stop building new test articles and spend more time at the drawing board refining the design and everything supporting it to ensure that the vehicle and operating procedures are well-thought-out enough to pass a few static fires and pressure tests without exploding. Testing to destruction typically follows basic design testing. At this point, I don't think they really understand what they're doing, unless they're intentionally trying to destroy things, in which case, mission accomplished.

tahanson43206 · 2020-05-29 20:57:47

For Louis re #692 ...

first, thanks for the link!

Second, who do you have in mind as "authorities" ???

I was impressed that the test site seemed intact upwind of the explosion.

The audio sounded like the camera operator answering a question from the Range Officer, confirming the camera operator was alright.

In short, it seemed to me that the range was properly designed in anticipation of just such an event.

My interpretation of the event is guided somewhat by my concept of what the "authorities" might have thought, and by lack of complete knowledge, but on the evidence of the video (and the survival of the camera operator and the Range safety officer), they ** might ** have thought "Excellent".

I'm looking forward to reading GW Johnson's evaluation of the test.

kbd512 re #693 ... I doubt they ** intentionally** created the explosion to prove the efficacy of the test site << grin >>

(th)

GW Johnson · 2020-05-30 10:22:20

Well, I don't know the inboard details of which cryogen's tank is forward, and which tank is aft, or whether these prototypes use the "header tank" internally-nested small tank geometry shown in the 2016 and 2017 Musk presentations.

It's also hard to tell what is air moisture condensation around either cryogen, and what might be water deluge, which has to be there, too. All I know for sure is that there was a sudden release of both methane and LOX as liquids mixing and evaporating upon the ground, and the mixing vapors quite easily found ignition. Probably the methane flare, but it could have been anything electrical, too.

The explosion was fast enough to be a full shock-wave detonation. You can't really see that in the link video, but I saw multiple camera angles in the video of some Irish fellow named Manley right after it on youtube, and the shockwave was clearly shown by its air moisture condensation wave.

There seemed to be two explosions, the second smaller one triggered when the mass simultor crashed back into whatever equipment was right around the test stand. My guess (and guess it is) is that the propellant load/unload plumbing got smashed by the crashing mass simulator, and its released contents blew up.

There is no clue that I understand about what leaked the propellants. There is only the history of welds not holding.

LOX is not methane, it doesn't blow up upon mixing into air. Methane and kerosene do, but methane is super-volatile compared to kerosene, and with a far lower initiation energy. In that respect, it resembles hydrogen more than anything else. That kind of behavior it far outside Spacex's prior experiences.

Beyond that, yo no sabe.

GW

louis · 2020-05-30 11:41:28

Scott Manley is Scottish - there's a clue in the name! lol

GW Johnson wrote:

Well, I don't know the inboard details of which cryogen's tank is forward, and which tank is aft, or whether these prototypes use the "header tank" internally-nested small tank geometry shown in the 2016 and 2017 Musk presentations.
It's also hard to tell what is air moisture condensation around either cryogen, and what might be water deluge, which has to be there, too. All I know for sure is that there was a sudden release of both methane and LOX as liquids mixing and evaporating upon the ground, and the mixing vapors quite easily found ignition. Probably the methane flare, but it could have been anything electrical, too.
The explosion was fast enough to be a full shock-wave detonation. You can't really see that in the link video, but I saw multiple camera angles in the video of some Irish fellow named Manley right after it on youtube, and the shockwave was clearly shown by its air moisture condensation wave.
There seemed to be two explosions, the second smaller one triggered when the mass simultor crashed back into whatever equipment was right around the test stand. My guess (and guess it is) is that the propellant load/unload plumbing got smashed by the crashing mass simulator, and its released contents blew up.
There is no clue that I understand about what leaked the propellants. There is only the history of welds not holding.
LOX is not methane, it doesn't blow up upon mixing into air. Methane and kerosene do, but methane is super-volatile compared to kerosene, and with a far lower initiation energy. In that respect, it resembles hydrogen more than anything else. That kind of behavior it far outside Spacex's prior experiences.
Beyond that, yo no sabe.
GW

GW Johnson · 2020-05-30 14:07:07

Sorry, Louis, from this side of the pond, and with my ear set to Texas drawl, Scottish and Irish accents sound just about the same to me.

I'm sure all us Americans sound about the same to you Brits, but our northern friends sound quite different from us, to our ears. It's a local thing, I'm sure.

GW

louis · 2020-05-30 17:33:56

Of course, I am sure it works the other way too! I don't think most Brits could distinguish a Canadian accent from an American one, let alone regional American accents.

I think in that old film How Green Was My Valley , they had the Welsh speaking Irish as well! lol

Actually there are parts of the UK where Scottish accents shade into Irish...I've sometimes thought I was hearing the Hebridean (Western Scottish Isles) accent but it has been Irish.

I had some Welsh ancestors who went to Texas...but some came back...felt homesick. Perhaps they couldn't understand the Texan drawl! lol

GW Johnson wrote:

Sorry, Louis, from this side of the pond, and with my ear set to Texas drawl, Scottish and Irish accents sound just about the same to me.
I'm sure all us Americans sound about the same to you Brits, but our northern friends sound quite different from us, to our ears. It's a local thing, I'm sure.
GW

louis · 2020-05-30 17:37:25

Of course, I am sure it works the other way too! I don't think most Brits could distinguish a Canadian accent from an American one, let alone regional American accents.

I think in that old film How Green Was My Valley , they had the Welsh speaking Irish as well! lol

Actually there are parts of the UK where Scottish accents shade into Irish...I've sometimes thought I was hearing the Hebridean (Western Scottish Isles) accent but it has been Irish.

I had some Welsh ancestors who went to Texas...but some came back...felt homesick. Perhaps they couldn't understand the Texan drawl! lol

GW Johnson wrote:

Sorry, Louis, from this side of the pond, and with my ear set to Texas drawl, Scottish and Irish accents sound just about the same to me.
I'm sure all us Americans sound about the same to you Brits, but our northern friends sound quite different from us, to our ears. It's a local thing, I'm sure.
GW

Void · 2020-05-30 18:26:50

Maybe I will get a spanking, but I am amused.

Although I am not that much lowland Scottish, they had a perfectly legitimate form of English, as Anglo-Saxons were there as well.
Just for fun, Scott Manley does not have spoken proper Scottish English.
https://www.bing.com/videos/search?q=Sc … &FORM=VIRE

It is hard for me not to be amused, as I am Swedish and Scottish among other things, as per genetic heritage.

I will cease to profane this thread with this off topic stuff. A little smile

Last edited by Void (2020-05-30 18:29:27)

New Mars Forums

Announcement

#676 2020-05-12 21:29:46

Re: Starship is Go...

#677 2020-05-13 06:06:15

Re: Starship is Go...

#678 2020-05-13 07:12:55

Re: Starship is Go...

#679 2020-05-13 11:15:53

Re: Starship is Go...

#680 2020-05-13 11:45:44

Re: Starship is Go...

#681 2020-05-15 11:41:51

Re: Starship is Go...

#682 2020-05-19 16:43:13

Re: Starship is Go...

#683 2020-05-21 11:41:59

Re: Starship is Go...

#684 2020-05-21 16:50:42

Re: Starship is Go...

#685 2020-05-23 13:17:56

Re: Starship is Go...

#686 2020-05-25 14:54:08

Re: Starship is Go...

#687 2020-05-25 19:39:02

Re: Starship is Go...

#688 2020-05-25 19:40:50

Re: Starship is Go...

#689 2020-05-26 11:35:50

Re: Starship is Go...

#690 2020-05-26 16:42:51

Re: Starship is Go...

#691 2020-05-26 17:30:02

Re: Starship is Go...

#692 2020-05-29 18:51:03

Re: Starship is Go...

#693 2020-05-29 20:00:49

Re: Starship is Go...

#694 2020-05-29 20:57:47

Re: Starship is Go...

#695 2020-05-30 10:22:20

Re: Starship is Go...

#696 2020-05-30 11:41:28

Re: Starship is Go...

#697 2020-05-30 14:07:07

Re: Starship is Go...

#698 2020-05-30 17:33:56

Re: Starship is Go...

#699 2020-05-30 17:37:25

Re: Starship is Go...

#700 2020-05-30 18:26:50

Re: Starship is Go...

Board footer