Author Topic: Discussion of NASA data review delaying SpaceX launches  (Read 34080 times)

Online meekGee

  • Senior Member
  • *****
  • Posts: 14517
  • N. California
  • Liked: 14469
  • Likes Given: 1408
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #60 on: 10/12/2023 10:27 pm »

In the good old days, the delay was no big deal, so the silliness was overlooked.  But now with bi-weekly launches, it's becoming apparent.

You have no place to call it silliness.
Your day will come.

nor is Starlink is not hurting for launch opportunities.
I call it as I see it.

And you still haven't, in any post, explained how the postponement is helping Psyche.

The closest thing you've said (to the truth) is that SpaceX signed up for this practice when they entered the contract, and it's a legal matter, got nothing to do with what's good for any specific flight.

The very formal version of "this is how it's done".
ABCD - Always Be Counting Down

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 37706
  • Cape Canaveral Spaceport
  • Liked: 21832
  • Likes Given: 430
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #61 on: 10/12/2023 10:52 pm »
I call it as I see it.


Yes, through SpaceX tinted glasses.

Offline mn

  • Full Member
  • ****
  • Posts: 1097
  • United States
  • Liked: 985
  • Likes Given: 354
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #62 on: 10/13/2023 12:06 am »
But that's true for all flights. What's special about the most recent one? Why isn't it just another data point?

Because it is new data

I do agree that NASA's rules are not the best for adapting to the current environment. However I do agree with Jim. Any flight is new data that may reveal a problem. If that new data is not reviewed prior to the next flight then if something goes bad then there is going to be a lot of egg of face, soul searching, etc.

However this is an approach that NASA has learnt the hard way, and to be honest, I see no reason why NASA shouldn't continue this way, until such a time that this approach can be properly reviewed and risk understood.

Bashing Jim is not really called for. I know he can be blunt, and very short, but lets assume that he has a tonne of insight that he cannot share and that insight supports NASA's approach. And as others have said, NASA is way more risk adverse than SpaceX with starlink flights.

Let's stop the antagonism towards Jim and take what he says at face value please.

Thank you for confirming that the only reason to delay is to avoid egg in the face.

As long as you agree that it doesn't actually improve anything...

Offline Comga

  • Senior Member
  • *****
  • Posts: 6494
  • Liked: 4603
  • Likes Given: 5289
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #63 on: 10/13/2023 01:37 am »

Of course there is “some data”. Starlink Group 6-22 in orbit is some data. B1067 sitting on ASOG is some data.

Not relevant.   

January 24, 1985 STS 51-C worst SRB o-ring blow by seen.

This is an argument against what might be a rational rule.
NASA has an entire year to analyze 51-C but didn’t find the danger that looks so obvious in hindsight, as do most failures.
WHY NASA missed it is the subject of several books and process improvements, but to some the 51-C SRB issues were not under scrutiny.
If something fails after 200+ sequential success, it might also be unscrutinized.  It doesn’t seem like it could be systemic.  One more data point won’t change a statistical distribution.
That will only be obvious in hindsight.

Quote
Many times I can't reveal what I know

Of course, but is this one of those times?
« Last Edit: 10/13/2023 01:38 am by Comga »
What kind of wastrels would dump a perfectly good booster in the ocean after just one use?

Offline matthewkantar

  • Senior Member
  • *****
  • Posts: 2157
  • Liked: 2595
  • Likes Given: 2287
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #64 on: 10/13/2023 01:51 am »
So why not just review 7-4 or 6-21? 

This is like Xeno's paradox.

More like Schrödinger‘s launch.

Or the ostrich with its head in the sand?

Delaying a commercial launch to cover NASA butts is lame.

There are some failures that are so obvious they need no analysis to require action.

If the Starlink launch is going to blow up on the pad, it could save the Psyche mission. Instead Psyche is in a position to save a run of the mill Starlink flight. 

Online zubenelgenubi

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 11691
  • Arc to Arcturus, then Spike to Spica
  • Sometimes it feels like Trantor in the time of Hari Seldon
  • Liked: 7810
  • Likes Given: 76187
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #65 on: 10/13/2023 02:33 am »
Observations from a fellow member:
Jim is laconic.  Like the Spartan reply to Philip of Macedon.  He's not a loquacious chap, at least on-line.

First of all, this splinter thread should not be a take-all-comers "rasslin'" match.  If you are not happy with his replies, then maybe someone should further research and write an NSF article about it. ✍️ Jim might consent to be a source?

Please also avoid turning this discussion into a Jim bear-baiting. 🙏 They are illegal, including here in the forum. 🐻🐻‍❄️  It's happened before, and the results are deleted.

[Philip: If I invade Laconia, I shall turn you out.
Spartan ephors' reply: If.]
Support your local planetarium! (COVID-panic and forward: Now more than ever.) My current avatar is saying "i wants to go uppies!" Yes, there are God-given rights. Do you wish to gainsay the Declaration of Independence?

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3436
  • Liked: 6221
  • Likes Given: 873
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #66 on: 10/13/2023 04:26 am »
Prevents from have to go before Congress, when Psyche buys it and there was problem in the previous launch but the data wasn't reviewed.
I think this cuts to the heart of the problem.

TECHNICALLY, the best strategy would be to launch Starlink, review what data you can, then (if things seem OK) launch Psyche.  If there is some horrible latent bug in F9, another launch is another chance to find it. If there is some subtle bug, there is still a better chance of finding it in the data by launching F9 first, as it can't possibly be found in time to help if the launch is delayed until after Psyche.

POLITICALLY, delaying the F9 launch avoids possible embarrassing questions if Psyche fails.   This may seem stupid to a rational person, but the target audience is not rational people, it's congressmen.  Remember when our President said "If we didn't test so much, we would not have so many cases!".  That's exactly the same attitude.


Offline steveleach

  • Senior Member
  • *****
  • Posts: 2362
  • Liked: 2921
  • Likes Given: 1006
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #67 on: 10/13/2023 08:06 am »
the launch vehicles are not static.  These are not the same design as a year ago.  Every upperstage is new.
I get that launches of newer designs need to be analysed, that makes total sense to me.  If you've had two hundred launches of version 1.0, then one launch of v1.1 and yours is going to be the second launch on v1.1 then there's clearly new information to be had.

But I'm still not at all sure I understand how that explains why you'd delay a launch so that it's not the most (chronologically) recent before yours. If you go ahead with the n-1 launch then you haven't lost the data from the n-2 launch, which is what you were going to be analysing anyway.

Anyway, thanks for taking the time to try to explain it me.

Offline edzieba

  • Virtual Realist
  • Senior Member
  • *****
  • Posts: 6391
  • United Kingdom
  • Liked: 9746
  • Likes Given: 41
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #68 on: 10/13/2023 10:47 am »
SpaceX are constantly changing not just vehicle designs, but across-the-board processes. Process changes have leave to failures, and even killed vehicles, and in one cost lost a mission. "Well the last launch worked, so everything must be fine" is not a useful standard for confirming system functionality.

AMOS-7 was lost because a process change (LHe load sequencing and load speed) uncovered a new failure mode (solid LOX formation within CoPV weave) in the upper stage. That was not the first launch to use that process change, and there were opportunities to review past load sequences and pick up fibre break noise from the accelerometer telemetry just as it was picked up for AMOS-7. SpaceX chose not to do that sort of detailed review (because that would be a lot of overhead and eat into their cost margins and reduce flight rates) and in this event it bit them.
Crew dragon C204 was lost because of a ground handling process issue, to the point a vehicle design change was deemed the best solution.
GPS III-04 aborted at T-2 and several M1Ds (including one on Crew-1) were replaced across the fleet because of a change in masking procedure for anodization of an internal engine component left lacquer residue that clogged a valve.
Starlink 5 v1.0 lost a core due to an engine failure and early shutdown (late enough for the vehicle to still make the target orbit), because a cleaning process left IPA in a fluid line that subsequently ignited in flight.

SpaceX choose not to perform detailed reviews after each launch, because they're willing to shoulder more risk. They can build more Starlinks, and even customer payloads are insured. NASA do not have this willingness, both for crew launches and for unique science missions that may not have budget to be replicated or may not have another launch window. So NASA looks at the most recent launch, reviews all the data they can get, to try and minimise the change of a new edge-case resulting in a mission failure.

NASA pay extra to SpaceX (over and above the normal commercial launch cost) for these sorts of extra reviews for their missions, and they accept contracts for launches with these conditions known in advance. If NASA is willing to pay for it, and SpaceX is willing to do it, what is exactly the problem?

Offline mandrewa

  • Full Member
  • ****
  • Posts: 629
  • Liked: 466
  • Likes Given: 8511
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #69 on: 10/13/2023 11:58 am »
AMOS-7 was lost because a process change (LHe load sequencing and load speed) uncovered a new failure mode (solid LOX formation within CoPV weave) in the upper stage.

I thought AMOS-6 AMOS-7 was lost during a test, not a launch.  There were exploring a faster LOX load sequence.  It wasn't the LOX load sequence they had been using in previous launches or that they planned to use in the AMOS-6 AMOS-7 launch.  It was a new data point.

I'm imaging a graph with one axis being fill time and another axis being perhaps pressure and yet another being temperature.  This was just a combination of conditions that they hadn't tried before.

Quote
That was not the first launch to use that process change, and there were opportunities to review past load sequences and pick up fibre break noise from the accelerometer telemetry just as it was picked up for AMOS-7.

If I understand you, you're saying that after the AMOS-6 AMOS-7 accident, they discovered that they could detect "fibre break noise" in the right circumstances.  But I'm guessing that they were not looking to detect any such thing before the accident.

I don't think they should be blamed for not looking for "fibre break noise" in earlier LOX loadings.  I suspect it's only because of the accident that we know that this is something to look for.
« Last Edit: 10/13/2023 12:15 pm by mandrewa »

Offline Brigantine

  • Full Member
  • ***
  • Posts: 303
  • NZ
  • Liked: 146
  • Likes Given: 445
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #70 on: 10/13/2023 12:02 pm »
This feels like game theory. [I'm not an expert, no inside knowledge, I'm just processing what I've read here and guessing]


If n-1 hasn't launched yet, then the expected value of n to NASA is V1 > (Contract Price + Opportunity Cost of a future mission)
p is the probability that data from launch n-1 (as assessed after full review) does not drop the expected value of launch n (or at least not materially)
V1 is a combination of V2*p + V3*(1-p)
V2 > V1, V3 < (Contract Price + Opportunity Cost of a future mission)

If n-1 doesn't launch, NASA's best play is to go ahead and launch n
If n-1 launches shortly before n is due to launch, then NASA's best play (to maximize their own value) is to delay n and look for excuses to cancel the contract.
If they find an excuse, they cancel. If they don't find an excuse, they increase their own value by a windfall gain (and SpaceX doesn't benefit)

SpaceX could just launch n-1, and let n be delayed at NASA's request - but then they would be in the same bind (whether to launch n+1) but with less launch window left.
So their best play is to deny NASA the option of delaying - i.e. withhold n-1. Don't show your hand too early, keep them guessing.
« Last Edit: 10/13/2023 12:05 pm by Brigantine »

Offline spacenut

  • Senior Member
  • *****
  • Posts: 5207
  • East Alabama
  • Liked: 2594
  • Likes Given: 2910
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #71 on: 10/13/2023 12:04 pm »
Amos 7 was years ago.  SpaceX has launched over 200 since then without problems.  This last minute review smells of politics. 

Online ZachS09

  • Space Savant
  • Senior Member
  • *****
  • Posts: 8474
  • Roanoke, TX
  • Liked: 2392
  • Likes Given: 2087
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #72 on: 10/13/2023 12:06 pm »
AMOS-7 was lost because a process change (LHe load sequencing and load speed) uncovered a new failure mode (solid LOX formation within CoPV weave) in the upper stage.

I thought AMOS-7 was lost during a test, not a launch.  There were exploring a faster LOX load sequence.  It wasn't the LOX load sequence they had been using in previous launches or that they planned to use in the AMOS-7 launch.  It was a new data point.

I'm imaging a graph with one axis being fill time and another axis being perhaps pressure and yet another being temperature.  This was just a combination of conditions that they hadn't tried before.

Quote
That was not the first launch to use that process change, and there were opportunities to review past load sequences and pick up fibre break noise from the accelerometer telemetry just as it was picked up for AMOS-7.

If I understand you, you're saying that after the AMOS-7 accident, they discovered that they could detect "fibre break noise" in the right circumstances.  But I'm guessing that they were not looking to detect any such thing before the accident.

I don't think they should be blamed for not looking for "fibre break noise" in earlier LOX loadings.  I suspect it's only because of the accident that we know that this is something to look for.

I didn’t know there was an AMOS 7. How come Spacecom didn’t make that satellite as a backup for AMOS 6, but rather mount it on Falcon 9 only for it to be lost in the test failure?
« Last Edit: 10/13/2023 12:07 pm by ZachS09 »
Liftoff for St. Jude's! Go Dragon, Go Falcon, Godspeed Inspiration4!

Offline spacenut

  • Senior Member
  • *****
  • Posts: 5207
  • East Alabama
  • Liked: 2594
  • Likes Given: 2910
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #73 on: 10/13/2023 12:21 pm »
Bureaucracies are a weight around advancement and innovation.  The Soviet Union crashed due to an overbearing weight of bureau's.   

Offline CorvusCorax

  • Full Member
  • ****
  • Posts: 1902
  • Germany
  • Liked: 4109
  • Likes Given: 2810
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #74 on: 10/13/2023 12:28 pm »
AMOS-7 was lost because a process change (LHe load sequencing and load speed) uncovered a new failure mode (solid LOX formation within CoPV weave) in the upper stage.

I thought AMOS-7 was lost during a test, not a launch.  There were exploring a faster LOX load sequence.  It wasn't the LOX load sequence they had been using in previous launches or that they planned to use in the AMOS-7 launch.  It was a new data point.

I'm imaging a graph with one axis being fill time and another axis being perhaps pressure and yet another being temperature.  This was just a combination of conditions that they hadn't tried before.

Quote
That was not the first launch to use that process change, and there were opportunities to review past load sequences and pick up fibre break noise from the accelerometer telemetry just as it was picked up for AMOS-7.

If I understand you, you're saying that after the AMOS-7 accident, they discovered that they could detect "fibre break noise" in the right circumstances.  But I'm guessing that they were not looking to detect any such thing before the accident.

I don't think they should be blamed for not looking for "fibre break noise" in earlier LOX loadings.  I suspect it's only because of the accident that we know that this is something to look for.

I didn’t know there was an AMOS 7. How come Spacecom didn’t make that satellite as a backup for AMOS 6, but rather mount it on Falcon 9 only for it to be lost in the test failure?

AMOS-7 was originally launched as AsiaSat 8 on a F9 v 1.1 on on 5 August 2014 - In 2017 the sat was leased to Spacecom and renamed AMOS-7, The sat is in orbit and operational.

https://en.wikipedia.org/wiki/AsiaSat_8

Obviously the post meant the ill-fated AMOS-6

Offline mn

  • Full Member
  • ****
  • Posts: 1097
  • United States
  • Liked: 985
  • Likes Given: 354
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #75 on: 10/13/2023 01:27 pm »
.......
what is exactly the problem?

The problem is that delaying the launch denies NASA data rather than giving them the data you just explained they need.

Let's say you are right and the Psyche vehicle has a new recent change.

Now if 6-22 does not have that change then no difference when it launches (and therefore no reason to delay it)

If hypothetically 6-22 has the same change, now you tell me which would you launch first?

Now of course any future launch could have the same change, so let's never launch Psyche

Offline mn

  • Full Member
  • ****
  • Posts: 1097
  • United States
  • Liked: 985
  • Likes Given: 354
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #76 on: 10/13/2023 03:25 pm »
Just to complete this thread: Psyche has now launched successfully and is on it's way.

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2531383#msg2531383

I hope we can put this thread to rest now. (until next time....)

Offline Tommyboy

  • Full Member
  • ***
  • Posts: 308
  • The Netherlands
  • Liked: 374
  • Likes Given: 598
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #77 on: 10/13/2023 03:43 pm »
Just to complete this thread: Psyche has now launched successfully and is on it's way.

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2531383#msg2531383

I hope we can put this thread to rest now. (until next time....)
Why? We still have no answer how holding back the Starlink launch could/might have affected the chance of success (or failure is your glass half empty) of the Psyche launch.

Offline mn

  • Full Member
  • ****
  • Posts: 1097
  • United States
  • Liked: 985
  • Likes Given: 354
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #78 on: 10/13/2023 03:57 pm »
Just to complete this thread: Psyche has now launched successfully and is on it's way.

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2531383#msg2531383

I hope we can put this thread to rest now. (until next time....)
Why? We still have no answer how holding back the Starlink launch could/might have affected the chance of success (or failure is your glass half empty) of the Psyche launch.

True, but everyone already said their opinion, sometimes you got to let it go.

Offline jimvela

  • Member
  • Full Member
  • ****
  • Posts: 1665
  • Liked: 917
  • Likes Given: 73
Re: Discussion of NASA data review delaying SpaceX launches
« Reply #79 on: 10/13/2023 04:12 pm »
Just to complete this thread: Psyche has now launched successfully and is on it's way.

https://forum.nasaspaceflight.com/index.php?topic=50260.msg2531383#msg2531383

I hope we can put this thread to rest now. (until next time....)
Why? We still have no answer how holding back the Starlink launch could/might have affected the chance of success (or failure is your glass half empty) of the Psyche launch.

The thread can't formally close until after the starlink launch. 

It would have been a bummer hypothetical if the starlink launch had failed, and then PSYCHE has to decide whether or not to proceed as they have only a limited window.

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1