Author Topic: F9 Block 5 Updates and Discussion  (Read 322985 times)

Offline AS-503

  • Full Member
  • ****
  • Posts: 431
  • Orion Fab Team
  • Colorado USA
  • Liked: 163
  • Likes Given: 116
Re: F9 Block 5 Updates and Discussion
« Reply #760 on: 05/11/2018 02:46 pm »
40% margin is not a requirement

Jim, I thought 1.4 structure margins were a requirement for manned rating.

You are showing requirements for pressurized vessels and not structures.

The second item in the table is "Metallic Propellant Tanks that are Pressurized Structures".
Doesn't this make up virtually the entire vehicle (booster and upper stage)?

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32546
  • Cape Canaveral Spaceport
  • Liked: 11333
  • Likes Given: 334
Re: F9 Block 5 Updates and Discussion
« Reply #761 on: 05/11/2018 02:47 pm »
F9 can take a single engine out at any time in the mission and still complete the mission.

No, it can't.  It flies 60-70% of the flight on one engine
« Last Edit: 05/11/2018 02:47 pm by Jim »

Offline lonestriker

  • Member
  • Posts: 89
  • Houston We Have A Problem
  • Liked: 106
  • Likes Given: 156
Re: F9 Block 5 Updates and Discussion
« Reply #762 on: 05/11/2018 03:03 pm »
F9 can take a single engine out at any time in the mission and still complete the mission.

No, it can't.  It flies 60-70% of the flight on one engine

Touchť :-). That was more likely me just misquoting Musk. I think he said booster phase actually.

He did pivot from his 40% margin on stress on the vehicle rambling into the weight and engine reliability info. What is generally covered by the "40% margin" for human rating that Elon was talking about?

Online Jakusb

  • Full Member
  • ****
  • Posts: 954
  • NL
  • Liked: 727
  • Likes Given: 405
Re: F9 Block 5 Updates and Discussion
« Reply #763 on: 05/11/2018 03:14 pm »
F9 can take a single engine out at any time in the mission and still complete the mission.

No, it can't.  It flies 60-70% of the flight on one engine

Depends on what is defined as Mission....
I guess Elon's remark is relevant for the payload's mission... Not so much the booster's mission of returning in one piece... ;)

Offline abaddon

  • Full Member
  • ****
  • Posts: 1747
  • Liked: 1228
  • Likes Given: 1123
Re: F9 Block 5 Updates and Discussion
« Reply #764 on: 05/11/2018 03:21 pm »
Depends on what is defined as Mission....
I guess Elon's remark is relevant for the payload's mission... Not so much the booster's mission of returning in one piece... ;)
Jim is referring to the second stage, which has only one engine and is obviously mission critical.

Offline matthewkantar

  • Full Member
  • ****
  • Posts: 993
  • Liked: 754
  • Likes Given: 806
Re: F9 Block 5 Updates and Discussion
« Reply #765 on: 05/11/2018 03:39 pm »
There are about 1400 seconds of engine run time covered by engine out capability, 600 seconds or less not covered. So 60-70% is in error. Also compare to competitors who offer exactly zero seconds of engine out coverage.

Matthew

Offline gongora

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 4300
  • US
  • Liked: 3827
  • Likes Given: 2186
Re: F9 Block 5 Updates and Discussion
« Reply #766 on: 05/11/2018 03:42 pm »
There are about 1400 seconds of engine run time covered by engine out capability, 600 seconds or less not covered. So 60-70% is in error. Also compare to competitors who offer exactly zero seconds of engine out coverage.

Matthew

Multiplying the first stage firing time by the number of engines is completely irrelevant.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32546
  • Cape Canaveral Spaceport
  • Liked: 11333
  • Likes Given: 334
Re: F9 Block 5 Updates and Discussion
« Reply #767 on: 05/11/2018 03:57 pm »
There are about 1400 seconds of engine run time covered by engine out capability, 600 seconds or less not covered. So 60-70% is in error. Also compare to competitors who offer exactly zero seconds of engine out coverage.

Matthew

wrong, total engine run time is meaningless, especially in light of quotes like this:

F9 can take a single engine out at any time in the mission and still complete the mission.
« Last Edit: 05/11/2018 03:57 pm by Jim »

Online hkultala

  • Full Member
  • ****
  • Posts: 902
  • Liked: 376
  • Likes Given: 362
Re: F9 Block 5 Updates and Discussion
« Reply #768 on: 05/11/2018 04:13 pm »
There are about 1400 seconds of engine run time covered by engine out capability, 600 seconds or less not covered. So 60-70% is in error. Also compare to competitors who offer exactly zero seconds of engine out coverage.

Matthew

Multiplying the first stage firing time by the number of engines is completely irrelevant.

No, it's not.

Assuming each engine failure probability of X per second, and this is the same for both stage engines.

The first stage burns something like 130 seconds during ascend on recoverable mode. Probability of at least one engine failing during ascend is about 1 - ((1-X)^ (130*9)).

The second stage burns about 360 seconds. Probability of failing during the first stage ascend is about 1 - ((1-X)^360).

Probability of failure in either is about 1- ((1-X)^(360+130*9)

Lets use value 0.00001 for X and then calculate some numbers. Then the probability of first stage engine failure is about 0.011 , 1.1 %, the probability of second stage failure is about 0.0036 , 0.36 %, and probability of failure in either is
0.0152, about 1.52 %.

(this my 0.00001 value is in correct range as they have now launched over 50 times and had one first stage engine failure and zero second stage engine failures).

So, probability of an engine failure is many times higher during the first stage ascend than during the second stage, even though the first stage burns for shorter time. So, of all the total expected engines failures, only about 0.0036 / 0.015 = 23% are expected to be second stage engine failures which means LOM.

edit:

But the most relevant comparison is to hypothetical similar launcher with just single first stage engine, and compare the LOM rates.

Then with the same 0.00001/second failure rate the total engine failure rate would only be 0.0049, but these would all result a LOM.

So multiple small engines with engine-out capability drops the LOM rate (of engine fail) only slightly, from 0.0049 to 0.0036 with this 0.00001/second reliability, compared to one big engine.

But compared to "few quite big engines without engine-out" (like proton or antares), the many small + engine out gives much better expected reliability.


« Last Edit: 05/11/2018 04:29 pm by hkultala »

Offline Rabidpanda

  • Full Member
  • ****
  • Posts: 527
  • Liked: 119
  • Likes Given: 572
Re: F9 Block 5 Updates and Discussion
« Reply #769 on: 05/11/2018 04:25 pm »
The 40% number sounds like a pretty clear reference to structural margin. Iím guessing that on previous versions of F9 the majority of the structure was designed to something like 1.25 ultimate FoS. And now for human rating they are designing to 1.4 ultimate FoS. Of course, this would not necessarily apply to all types of structure, because different types of structure/components have different margin requirements.

As for any kind of fault tolerance / reliability requirements (which engine-out would fall under) that would be completely different from a 40% structural margin.

Offline matthewkantar

  • Full Member
  • ****
  • Posts: 993
  • Liked: 754
  • Likes Given: 806
Re: F9 Block 5 Updates and Discussion
« Reply #770 on: 05/11/2018 04:54 pm »
There are about 1400 seconds of engine run time covered by engine out capability, 600 seconds or less not covered. So 60-70% is in error. Also compare to competitors who offer exactly zero seconds of engine out coverage.

Matthew

wrong, total engine run time is meaningless, especially in light of quotes like this:

F9 can take a single engine out at any time in the mission and still complete the mission.

Wrong

Offline chrisking0997

  • Full Member
  • ***
  • Posts: 301
  • NASA Langley
  • Liked: 104
  • Likes Given: 240
Re: F9 Block 5 Updates and Discussion
« Reply #771 on: 05/11/2018 04:55 pm »
Is there some reason we (well, not we but...) are not taking the "engine out" quote to obviously refer to first stage flight only?   I get that the second half of the quote says "the mission", but I dont see how anyone could misinterpret that to mean that second stage engine failure is somehow going to be mitigated by any of the engines of a first stage that has been or will be shortly staged.  I suppose wording matters, but if thats the only point here I think we get it
Tried to tell you, we did.  Listen, you did not.  Now, screwed we all are.

Offline Comga

  • Senior Member
  • *****
  • Posts: 4497
  • Liked: 1746
  • Likes Given: 1446
Re: F9 Block 5 Updates and Discussion
« Reply #772 on: 05/11/2018 05:00 pm »
Of the non-technical details I liked this:
Quote
This is arguably Falcon 9 version 6, in sort of normal vernacular. Because we had version one, version 1.1 which was really like version 2, arguably a version in between that, and then a bunch of blocks. So we should probably just go back, I'm sure the internet's already done this, and have a more sensible description of the versioning. But think of it as like, at least version 6 of the rocket.

Sounds like Musk is assigning us, as representatives of the internet, with redoing the "versioning"
"F9.6" anyone?
What kind of wastrels would dump a perfectly good booster in the ocean after just one use?

Offline Comga

  • Senior Member
  • *****
  • Posts: 4497
  • Liked: 1746
  • Likes Given: 1446
Re: F9 Block 5 Updates and Discussion
« Reply #773 on: 05/11/2018 05:13 pm »
Active water cooling of the base heat shield during reentry?!

Yeah that was perhaps the biggest surprise. But cool!  8)

Quote
But the reason I picked the heat shield, it's also a big improvement. And we replaced the old composite structure with a high- temperature titanium structure to support rapid reuse. The base heat shield will also somewhat actively cooled with water. So we're finding things some things are really just, during the very high-energy phases of re-entry, ascent does not require them, but during the high-energy phases of re-entry, where you have a hypersonic shock-shock impingement, it generates a very hot spot, and you kind of have to use a high-melting point material, a high-temperature material, plus active water cooling in certain places on the base of the heat shield.

This seems a non-trivial addition of subsystems and mass.  The first stage would have to have a water reservoir and plumbing to circulate it to the hot structure.  Really hope the details show up some day in an updated illustration of the Falcon 9.  It further demonstrates SpaceX's prioritization of reuse, and ultimately cost reduction, over payload maximization.

It could indicate what had to be done to the earlier block first stages to refly them.  Perhaps replacing the composite structure that got cooked.

edit: Although I wonder what Musk meant by "somewhat actively cooled".  What qualifies as semi-active water cooling?  Perhaps the water reservoir is in contact with the heat shield, and it just boils away through a vent without circulating.
« Last Edit: 05/11/2018 05:17 pm by Comga »
What kind of wastrels would dump a perfectly good booster in the ocean after just one use?

Offline Prettz

  • Full Member
  • **
  • Posts: 277
  • Atlanta, GA
  • Liked: 111
  • Likes Given: 17
Re: F9 Block 5 Updates and Discussion
« Reply #774 on: 05/11/2018 06:13 pm »
The 40% number sounds like a pretty clear reference to structural margin. Iím guessing that on previous versions of F9 the majority of the structure was designed to something like 1.25 ultimate FoS. And now for human rating they are designing to 1.4 ultimate FoS. Of course, this would not necessarily apply to all types of structure, because different types of structure/components have different margin requirements.

As for any kind of fault tolerance / reliability requirements (which engine-out would fall under) that would be completely different from a 40% structural margin.
No, SpaceX has always talked about how F9 is designed for 40% safety margin (or whatever term they used, I can't remember) rather than the "standard" 25%. That's not new.

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32546
  • Cape Canaveral Spaceport
  • Liked: 11333
  • Likes Given: 334
Re: F9 Block 5 Updates and Discussion
« Reply #775 on: 05/11/2018 06:33 pm »
There are about 1400 seconds of engine run time covered by engine out capability, 600 seconds or less not covered. So 60-70% is in error. Also compare to competitors who offer exactly zero seconds of engine out coverage.

Matthew

wrong, total engine run time is meaningless, especially in light of quotes like this:

F9 can take a single engine out at any time in the mission and still complete the mission.

Wrong

Just stop with the misinformation

Offline gongora

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 4300
  • US
  • Liked: 3827
  • Likes Given: 2186
Re: F9 Block 5 Updates and Discussion
« Reply #776 on: 05/11/2018 06:58 pm »
edit: Although I wonder what Musk meant by "somewhat actively cooled".  What qualifies as semi-active water cooling?  Perhaps the water reservoir is in contact with the heat shield, and it just boils away through a vent without circulating.

Sounds to me like "in some part" instead of "somewhat".  The audio is pretty bad.

Offline biosehnsucht

  • Full Member
  • ***
  • Posts: 322
  • Liked: 112
  • Likes Given: 258
Re: F9 Block 5 Updates and Discussion
« Reply #777 on: 05/11/2018 07:00 pm »
I feel like we've pulled a Bill Clinton and started arguing the definition of "is" is.

Offline IainMcClatchie

  • Full Member
  • ***
  • Posts: 344
  • San Francisco Bay Area
  • Liked: 220
  • Likes Given: 331
Re: F9 Block 5 Updates and Discussion
« Reply #778 on: 05/11/2018 07:03 pm »
Although I wonder what Musk meant by "somewhat actively cooled".  What qualifies as semi-active water cooling?  Perhaps the water reservoir is in contact with the heat shield, and it just boils away through a vent without circulating.

That's my guess.  The cooling is needed while under deceleration.  The titanium should be good for 300 to 400 degrees C, so my guess is that they have a high pressure vent (8 to 10 MPa).  I'd plumb that vent so that the steam goes back down to the bottom of the vehicle and flows over whatever surface they've seen getting the most heat damage.  If you get the water fairly hot before you release steam, you can get a lot of volume to inject into that boundary layer.

Offline gongora

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 4300
  • US
  • Liked: 3827
  • Likes Given: 2186
Re: F9 Block 5 Updates and Discussion
« Reply #779 on: 05/11/2018 08:00 pm »
Another transcript is available on the CNBC site (the full pdf is below a bunch of quotes).  Of course it's probably 95% the same as the transcript we already had, but a few words differ here and there.  (I actually ran into one sentence where each of the transcripts was missing a word that was contained in the other one, so neither of them 100% matches the audio.)

https://www.cnbc.com/2018/05/11/full-elon-musk-transcript-about-spacex-falcon-9-block-5.html

Tags: