Author Topic: SpaceX will almost certainly have another failure within the next 3 years.  (Read 77744 times)

Offline The Amazing Catstronaut

  • Full Member
  • ****
  • Posts: 1065
  • Arsia Mons, Mars, Sol IV, Inner Solar Solar System, Sol system.
  • Liked: 759
  • Likes Given: 626
The first Delta IV Heavy launch was a failure.

This directly indicates the importance of a related observation: not all failures "count" equally.

It really helps if for example your payload is called, "DemoSat." ;)

It also helps when your customer is kind enough to say just after the anomaly:
"We are very pleased with the overall performance of the Delta 4-Heavy Demo"

Those were the words of Col. John Insprucker, EELV program director at the Space and Missile Systems Center.
http://www.spaceflightnow.com/delta/d310/041222update.html

So for example if FH drops its "Demo Flight" payload into the drink, that just doesn't count against them as much as blowing up a customer's payload on the pad during a static fire.


It's the constant upgrades. Compare first D4H and the first F9 to a 2016 F9 and a 2016 D4H. SpaceX has a much higher upgrade and optimisation curve - but not every upgrade is an optimisation. Every performance and time benefit must stack against risk. Your great-granddaddy F9 is not your kid sister F9. It's like the analogy of an ancestral axe in which nearly every component has been replaced, but it's still the same axe. The fact that the F9 has always worn the name F9 causes us to think about it wrong - the first F9 is arguably closer in lineage to F1 than the (prospectively) final iteration of next year.

Then think about the value-to-loss of those upgrades. In the end, the prospective improvements of those upgrades stack against the relative losses. SpaceX views it as acceptable and necessary risk - that's their calculus. What's ours?
« Last Edit: 12/29/2016 12:03 pm by The Amazing Catstronaut »
Resident feline spaceflight expert. Knows nothing of value about human spaceflight.

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
This suggests a related question. If another F9 failure is likely, what can SpaceX do to mitigate impact?
Transparency on the MIB seems to help, along with a prompt RTF.

The problem (shared by all VTO TSTOs) is there does not seem to be anywhere you can fit an LES that gets the payload back safely.

Do that and the customers may grumble about schedule slippage, or the need to refly on another LV, but they won't be nearly as annoyed as a full LOM.

This is called insurance. 
For crew, no 'insurance' is sufficient for the priceless payload, so an LES is required.

Commercial companies make bets every day that may or may not work out -- ones that don't are accepted as a cost of doing business.  All of the 70 or so manifested launches are free to move on to another launch services company if and when the bet is no longer justifiable business-wise.  Another failure 50 launches from now will be met with little or no abandoning ship... a failure 5 launches from now might have a different response.
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline MP99

It's an interesting question whether the landing and inspection of used first stages will allow SpaceX to detect dodged bullets, and fix them before they cause a failure in flight.

OTOH, ISTM that the early days of reuse give further opportunities for failures.

Cheers, Martin

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39243
  • Minnesota
  • Liked: 25169
  • Likes Given: 12100
In principle reuse allows them to fix things. In fact, they already have learned from the cores they've recovered. But it's a new thing, and anything new has potential to introduce or uncover failure modes. I bet that it's not too bad, as they've tested a flown core a bunch at McGregor.
« Last Edit: 12/29/2016 02:13 pm by Robotbeat »
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39243
  • Minnesota
  • Liked: 25169
  • Likes Given: 12100
This suggests a related question. If another F9 failure is likely, what can SpaceX do to mitigate impact?
Transparency on the MIB seems to help, along with a prompt RTF.

The problem (shared by all VTO TSTOs) is there does not seem to be anywhere you can fit an LES that gets the payload back safely.

Do that and the customers may grumble about schedule slippage, or the need to refly on another LV, but they won't be nearly as annoyed as a full LOM.

This is called insurance. 
For crew, no 'insurance' is sufficient for the priceless payload, so an LES is required.

Commercial companies make bets every day that may or may not work out -- ones that don't are accepted as a cost of doing business.  All of the 70 or so manifested launches are free to move on to another launch services company if and when the bet is no longer justifiable business-wise.  Another failure 50 launches from now will be met with little or no abandoning ship... a failure 5 launches from now might have a different response.
This.

Insurance is a thing. It's why companies are fine launching even on Proton, which has like 88-92% reliability and isn't improving or likely to dramatically improve. Falcon 9, even with its failures, is no worse than Proton. Particularly if SpaceX simply decides not to mount payloads while they static fire the stage! :)

SpaceX has two reliability levels. About 96% if they never static fire the rocket with a payload on top again (this wasn't a flight failure, remember!), and 93% if they do. Even the latter is better than Proton's 89-92%.
« Last Edit: 12/29/2016 01:17 pm by Robotbeat »
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline muomega0

  • Full Member
  • ****
  • Posts: 862
  • Liked: 70
  • Likes Given: 1
By the way, I think ULA may have 99% company-wide launch reliability at this point. You could even make an okay case for 99.5%. But all that changes as soon as they field Vulcan or even a re-engined Atlas.

So suppose SpaceX DOES succeed in avoiding a failure of Falcon 9 during the next 3 years and gets that high launch rate. SpaceX will have proved their reliability, probably made significant profit by the end of 2019 (no doubt plowed back into development), at just the moment ULA needs to introduce a new rocket (Vulcan) with all the risk that entails.
This is just one of many reasons to justify depot centric:  the ability to launch dirt Cheap Class D propellant in common configurations for NASA BEO missions...flight rate where the propellant increases in value from pennies to many $/kg once in LEO...The risk is the cost of the LV and it allows incremental improvements as well....

A failure is no big deal.....its just the cost of the (reusable) LV...

Now compare these approaches with SLS/Orion....

The insurance companies would clearly see the advantage as well.

Offline MP99

In principle reuse allows them to fix things. In fact, they already have learned from the cores they've recovered. But it's a new thing, and anything new has potential to introduce or uncover failure modes. I bet that it's not too bad, as they've tested a flown core a bunch at McGregor.
[Includes RB's edit]
This puts the stage through pressurization and cryogenic cycles, and puts a lot of minutes on 9 different engines.

However, I wonder how close this comes to putting the airframe under flight loads, especially high G with u/s and payload on top, and dynamic loads from the engines?

Cheers, Martin
« Last Edit: 12/29/2016 05:31 pm by MP99 »

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39243
  • Minnesota
  • Liked: 25169
  • Likes Given: 12100
Doh, should be "to fix things" not "toxic things"
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39243
  • Minnesota
  • Liked: 25169
  • Likes Given: 12100
In principle reuse allows them toxic things. In fact, they already have learned from the cores they've recovered. But it's a new thing, and anything new has potential to introduce or uncover failure modes. I bet that it's not too bad, as they've tested a flown core a bunch at McGregor.
This puts the stage through pressurization and cryogenic cycles, and puts a lot of minutes on 9 different engines.

However, I wonder how close this comes to putting the airframe under flight loads, especially high G with u/s and payload on top, and dynamic loads from the engines?

Cheers, Martin
They had a fancy loading structure, remember, for the reuse testing. And they've run it through at least 7 mission cycles now, enough to prove that one reflight is still far from the far side of the bathtub curve.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
EM has also recently said that the core is good for an 'unlimited' number of launches... engines 40 or so between refurbishments.  Launching a few of these this year, even with a failure (one type of failure* that might not even affect flights on rest of fleet), goes a long way toward establishing the hardware for reuse if post-flight inspections match the design wear expectations.

*A FH staging failure is another type of failure that probably won't affect flight operations too greatly.
« Last Edit: 12/29/2016 02:30 pm by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39243
  • Minnesota
  • Liked: 25169
  • Likes Given: 12100
To be clear, the current cores they have are maybe suited to 10 reuses, but the later version of Falcon 9, due to arrive in 2017, will be made for more reuses.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline watermod

  • Full Member
  • ****
  • Posts: 519
  • Liked: 177
  • Likes Given: 153
I view SpaceX's experience as lessons learned for them and others.
As somebody outside looking in and trying to see what I would do in their place to avoid another failure I would see it this way.
1) Pressurization of the fuel is the largest current problem.
2) Oil based fuels can't be easily pressurized by themselves
3) That being the case and Merlin requiring that fuel... if you don't swap it out you are stuck with the light weight helium.
4) Liquid helium composite tanks in Lox being your biggest danger - go non-composite or get them out of the LOX tank and into something else? Or do what SpaceX is doing and try to change process.
So their major risk calculation must be process risk & cost vs design change risk & cost.
5) a corollary to all this is that SpaceX will need experience with their new methane engine(s).   The experimenter suggests an occasional second stage with a methane engine instead of the Merlin.  This gets around the the biggest second stage risk but not the first stage one and provides confidence for the next rocket.

Offline watermod

  • Full Member
  • ****
  • Posts: 519
  • Liked: 177
  • Likes Given: 153
The best place to experiment with the second stage is with the SpaceX network of sats.
Launch the first few launches with traditional 2nd stages to start R&D on the sat process then experiment with the second stages and later the firsts.   That way, if there are problems the payload loss is only SpaceX assembly line costs and not customer end product costs.

Offline old_sellsword

  • Full Member
  • ****
  • Posts: 632
  • Liked: 531
  • Likes Given: 470
In principle reuse allows them toxic things. In fact, they already have learned from the cores they've recovered. But it's a new thing, and anything new has potential to introduce or uncover failure modes. I bet that it's not too bad, as they've tested a flown core a bunch at McGregor.
This puts the stage through pressurization and cryogenic cycles, and puts a lot of minutes on 9 different engines.

However, I wonder how close this comes to putting the airframe under flight loads, especially high G with u/s and payload on top, and dynamic loads from the engines?

Cheers, Martin
They had a fancy loading structure, remember, for the reuse testing. And they've run it through at least 7 mission cycles now, enough to prove that one reflight is still far from the far side of the bathtub curve.

The orange cap on the top was not for simulating flight conditions, it was holding the stage down so it didn't rip itself out of the hold downs.

Offline Tonioroffo

  • Full Member
  • *
  • Posts: 131
  • Belgium
  • Liked: 83
  • Likes Given: 107
If SpaceX keeps testing a lot on F9H, reuse first stages, etc, there will be more RUD's.  But if the FAA keeps F9 grounded for 6 months after every incident, is it then realistic for SpaceX to keep being on the edge if innovation  instead of being conservative?  Sounds like being stuck between a rock and a Hard place to me.

Sent from my LG-H850 using Tapatalk


Offline FutureSpaceTourist

  • Global Moderator
  • Senior Member
  • *****
  • Posts: 47310
  • UK
    • Plan 28
  • Liked: 80116
  • Likes Given: 36283
Surely it's the complexity of the investigation and any corrective action that determines whether it's 6 months, not the FAA? Unless there's any evidence that the FAA are slowing the process down (other than the necessary approval at the end of the process of SpaceX's report and action closeouts)

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
If SpaceX keeps testing a lot on F9H, reuse first stages, etc, there will be more RUD's.  But if the FAA keeps F9 grounded for 6 months after every incident, is it then realistic for SpaceX to keep being on the edge if innovation  instead of being conservative?  Sounds like being stuck between a rock and a Hard place to me.

Sent from my LG-H850 using Tapatalk

This return to flight is slightly over four months after the incident.  If you recall the failure at McGregor during testing, no flight time was lost.  A 'failure' could produce a wide range of delays, depending on its applicability to other flight hardware.  SpaceX is on a path that departed from 'conservative' long ago, and is distancing itself all the more as Raptor, BFR, and all hit their stride.  There is no turning back, IMO.
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline sdsds

  • Senior Member
  • *****
  • Posts: 7037
  • “With peace and hope for all mankind.”
  • Seattle
  • Liked: 1909
  • Likes Given: 1889
SpaceX is on a path that departed from 'conservative' long ago, and is distancing itself all the more as Raptor, BFR, and all hit their stride.  There is no turning back, IMO.

Why do you see their future as a single path? Why one or the other? Why not a bifurcation where the company becomes both more conservative (with F9) and equally or more wild-and-crazy with Mars stuff?
— 𝐬𝐝𝐒𝐝𝐬 —

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39243
  • Minnesota
  • Liked: 25169
  • Likes Given: 12100
SpaceX is going on both paths simultaneously. They're nailing down F9 (shortly), it will be their workhorse with first stage reuse. Falcon Heavy for the few payloads that need something bigger.

The aggressive stuff will now be for Mars, not F9.

And by the way: innovation adds value even before it shows up in per-launch rate. Besides making SpaceX the most exciting place to work, it also adds value to the company. They own Merlin 1D and Falcon 9. And Raptor. And Dragon. Push the state of the art further, and you've now increased the value of the company.

Think of it this way: suppose that SpaceX has an IPO. What is it worth? Now, imagine SpaceX has an IPO but right before, they invent a warp drive. They don't launch any customer payloads or sign any contracts for it, they just demonstrate it conclusively. NOW if they have an IPO, they're worth far more. So it'd be stupid for SpaceX to stop innovation entirely while waiting for RTF or whathaveyou. The innovation adds future value all by itself.
« Last Edit: 12/29/2016 08:18 pm by Robotbeat »
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline AncientU

  • Senior Member
  • *****
  • Posts: 6257
  • Liked: 4164
  • Likes Given: 6078
SpaceX is on a path that departed from 'conservative' long ago, and is distancing itself all the more as Raptor, BFR, and all hit their stride.  There is no turning back, IMO.

Why do you see their future as a single path? Why one or the other? Why not a bifurcation where the company becomes both more conservative (with F9) and equally or more wild-and-crazy with Mars stuff?

Good point.  To me, F9 flying 'conservative' would mean expendable, no pushing Merlin, no sub-cooled fuel, no landing legs, grid fins, etc... just fly payloads and splash the hardware like every other conservative entity.  All the envelop pushing is for reuse and high tempo operations... all for the future.  Maybe an option is (was) to branch the launch business into a 'conservative' launch services department, but it seems that ship has sailed.
« Last Edit: 12/29/2016 09:44 pm by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Tags:
 

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