Author Topic: SpaceX Starship : First Flight : Starbase, TX : April 2023 - DISCUSSION THREAD 2  (Read 475895 times)

Offline ulm_atms

  • Rocket Junky
  • Full Member
  • ****
  • Posts: 921
  • To boldly go where no government has gone before.
  • Liked: 1550
  • Likes Given: 743
In fairness foresight on this one is 20-20 too. A flat concrete pad was an obvious problem. Multiple people in the industry told Elon and SpaceX this would be a problem. But they did it anyway.

I love the company but this was entirely avoidable and the conclusion completely obvious.

In equal fairness, multiple people in the industry told Elon not to even attempt Starship, Dragon, or Falcon 9, or try to make any of them reusable, or even to start SpaceX at all, and told him those things with far more credibility and historical basis than they're arguing this relatively new debate.

Every innovation ever was "entirely avoidable."  Let's all be glad someone at some time failed to avoid it.
Yea, so many people thing failing is 100% negative and should be avoided at all costs.  Failing and understanding why it failed is how most people learn.  Elon has his issues...but he has delivered a swift kick in the pants to the world wide space industry.  SS/SH will be the next kick IMO...It's real now and they can't pretend it doesn't exist anymore.

Now I feel the excitement that the people that lived through all of Appllo's dev and ops must of felt like.  I like it!  ;D

Online ugordan

  • Senior Member
  • *****
  • Posts: 8508
    • My mainly Cassini image gallery
  • Liked: 3520
  • Likes Given: 757
Not surprisingly the FAA grounds starship following the failure.
Not sure about the actual impact on future flights, but it seems the environmental damage was significant
https://www.politico.com/news/2023/04/20/spacex-starship-explode-elon-musk-00093042

 ??? I see no reference whatsoever to significant "environmental damage". A mishap investigation is standard procedure for FAA to require for this kind of thing.

Click-baity title, IMHO.


Offline deMangler

  • Member
  • Posts: 7
  • UK
  • Liked: 24
  • Likes Given: 344
I was quietly wondering why they made some design choices with the GSE and OLM.
Now it seems they weren't trying to minimise damage and maximise progress in the way I thought. It was more about maximising potential useful data and minimising overinvestment on unknowns.


Offline Stan-1967

  • Full Member
  • ****
  • Posts: 1123
  • Denver, Colorado
  • Liked: 1164
  • Likes Given: 610
They were literally one wayward concrete chunk from the entire orbital complex going up in a massive fireball. There's no way that you can look at what happened and not see that as a possible outcome.

Really?  Literally multiple millions of chunks of wayward concrete were blasted into the surrounding infrastructure,  but if there had been +1 more,   Disaster fireball ensues?  None of the millions of known chunks were wayward?  Just needed +1 more?


Offline eeergo

There was no stage separation or flip maneuver commanded.
The official SpaceX statement itself confirms this in that they state that the booster experienced multiple engine outs and then could no longer maintain attitude.
You may wish to zoom you screen in: the SpaceX statement says lost altitude, not attitude.

Flip start was commanded and expected (coinciding with commentators calling out start of separation, and camera switches timed to catch flip and separation, which then did not occur). Continued flipping was not intentional.

I had a conversation with a SpaceX engineer who corroborates this. The flip maneuver is apparently designed to be a pretty dramatic change in attitude that then reverses and swings the ship off the interstage just after MECO. HPU did not fail according to my source. TVC was operating fine during the flip maneuver, but he suspects that there was an anomaly with the software / interstage. Software got confused and the commands for the second half of the flip maneuver + MECO didn't execute as planned.

We literally saw a head-on shot of the HPU exploding, sending shrapnel flying everywhere, leaking a brightly-yellow-burning hydrocarbon into the exhaust stream for several seconds, and leaving a mighty hole in the stack where it was mounted.

A less clear event of the sorts happened shortly before "flipping" (either intentional or accidental), which many interpreted as the second, and last, SH HPU giving way too.

Regardless, the flip happened much earlier than BECO or separation was programmed for, both according to the published timeline and to live callouts. Even if the trajectory was off-nominal, both altitude and velocity were much lower than expected, so separation should if anything have been adjusted to happen later, at propellant depletion - not earlier. Furthermore, the separation couldn't have happened under booster thrust, no matter the attitude or angular rates. So in any case the first failure if we're to assume all the initial gyrations were intentional, was failure to effect BECO, not separation itself.

Are you sure no information was lost in your interview?

Offline chrisking0997

  • Full Member
  • ***
  • Posts: 355
  • NASA Langley
  • Liked: 127
  • Likes Given: 317
I've looked at many videos of the liftoff and it's clear to see that within only a few seconds of hold-down release that the stack was clearly non-vertical. Equally clear is that this was not nominal in any way because you can clearly see the vehicle struggling to compensate and regain the vertical, which it did. I have a few thoughts on why it was lifting at such an angle but at this point they are speculation.

makes you wonder if all the engines cleared the olm
Tried to tell you, we did.  Listen, you did not.  Now, screwed we all are.

Offline tenkendojo

  • Member
  • Posts: 97
  • usa
  • Liked: 90
  • Likes Given: 231
https://twitter.com/sciguyspace/status/1649381415442698242

Quote
I've spoken with half a dozen employees at SpaceX since the launch. If their reaction is anything to go by, the Starship test flight was a spectacular success. Of course there's a ton to learn, to fix, and to improve. It's all super hard work. But what's new? Progress is hard.
Hmm. the flip side to this statement is that they had pretty low expectations. Lower than mine, at least.

I've been closely following the development of SS for the whole time, and honestly I was expecting the 1st test launch will most likely ends up with the whole stack blowing up on the launch pad. I'm very impressed that SS managed to clear the launch pad at all.

Offline jimvela

  • Member
  • Full Member
  • ****
  • Posts: 1659
  • Liked: 891
  • Likes Given: 69
There was no stage separation or flip maneuver commanded.
The official SpaceX statement itself confirms this in that they state that the booster experienced multiple engine outs and then could no longer maintain attitude.
You may wish to zoom you screen in: the SpaceX statement says lost altitude, not attitude.

Flip start was commanded and expected (coinciding with commentators calling out start of separation, and camera switches timed to catch flip and separation, which then did not occur). Continued flipping was not intentional.

I had a conversation with a SpaceX engineer who corroborates this. The flip maneuver is apparently designed to be a pretty dramatic change in attitude that then reverses and swings the ship off the interstage just after MECO. HPU did not fail according to my source. TVC was operating fine during the flip maneuver, but he suspects that there was an anomaly with the software / interstage. Software got confused and the commands for the second half of the flip maneuver + MECO didn't execute as planned.

Tests that go bad can be a blessing.  It's why test often/test early is a good idea.   In that regard, there are going to be a HUGE number of lessons here for SpaceX to learn.

It appears that this novel separation implementation has a flaw in that if a starship has to abort from a booster, it may be unable to do so- or at least hampered by the separation system limitations.

If the clamps (or whatever retention mechanism exists) cannot be released by the starship (e.g. its avionics can directly actuate the mechanisms), then it's a bad day in an abort scenario.  Also a bad day if the booster engines wont/cant shut down in an abort. 

It doesn't appear that this launch had an abort commanded or perhaps not even possible for starship escape.  That seems like a pretty obvious shortcoming and omission from what might seem a minimum viable launch FSW implementation.  The worst impact is if that had been possible, this failed launch would have at least ticked off the abort scenario and provided extremely valuable data on how an abort works.  Maybe the capability was there and this is just exposing other bad bugs in the system. 

Even with the booster perhaps not releasing the clamps, It may still be possible to light up the engines on the starship to abort from a (literally) sideways launch, but even that hints at changes needed to the system. 
Perhaps blow-out panels or similar provisions on the booster interstage so that when the starship decides to get off the stack, it still has a fighting chance of not blowing its backside wide open from overpressure during engine start up.  Maybe that already exists- but I've not seen indications of it.

There's a picture in one of the threads from starship after the AFTS activation on the booster that appears to show the interstage still fully attached to the base of the starship.  That highlights the flaw in the implementation.  That starship would have had no chance of getting away from a failing booster with that flaw present.

There's mention upthread that perhaps a single booster center engine did shut down at around the time the flip/separation would have been occurring, and then relit shortly after.  Is that true?

Does anyone know where the flight computer(s) for the super heavy are?  I believe this system uses distributed avionics and that all the engine controllers are independently in communication with the flight computer(s).  Is it possible that there was severe damage to the interconnection in the base of the booster, resulting in commanding to the other elements of the booster being compromised?  [a question from ignorance, no assertion here...]  That would also potentially be a reason that separation did not occur.

Online chopsticks

  • Full Member
  • ****
  • Posts: 1059
  • Québec, Canada
  • Liked: 1076
  • Likes Given: 164
There was no stage separation or flip maneuver commanded.
The official SpaceX statement itself confirms this in that they state that the booster experienced multiple engine outs and then could no longer maintain attitude.
You may wish to zoom you screen in: the SpaceX statement says lost altitude, not attitude.

Flip start was commanded and expected (coinciding with commentators calling out start of separation, and camera switches timed to catch flip and separation, which then did not occur). Continued flipping was not intentional.

I had a conversation with a SpaceX engineer who corroborates this. The flip maneuver is apparently designed to be a pretty dramatic change in attitude that then reverses and swings the ship off the interstage just after MECO. HPU did not fail according to my source. TVC was operating fine during the flip maneuver, but he suspects that there was an anomaly with the software / interstage. Software got confused and the commands for the second half of the flip maneuver + MECO didn't execute as planned.

We literally saw a head-on shot of the HPU exploding, sending shrapnel flying everywhere, leaking a brightly-yellow-burning hydrocarbon into the exhaust stream for several seconds, and leaving a mighty hole in the stack where it was mounted.

A less clear event of the sorts happened shortly before "flipping" (either intentional or accidental), which many interpreted as the second, and last, SH HPU giving way too.

Regardless, the flip happened much earlier than BECO or separation was programmed for, both according to the published timeline and to live callouts. Even if the trajectory was off-nominal, both altitude and velocity were much lower than expected, so separation should if anything have been adjusted to happen later, at propellant depletion - not earlier. Furthermore, the separation couldn't have happened under booster thrust, no matter the attitude or angular rates. So in any case the first failure if we're to assume all the initial gyrations were intentional, was failure to effect BECO, not separation itself.

Are you sure no information was lost in your interview?

I agree that we saw one of the HPUs explode shortly after liftoff, but now I'm not so sure that the other one failed.

Look at the end of Micheal Baylor's video. You can clearly see the centre engines gimballing while the other engines' plume remains straight.

https://twitter.com/nextspaceflight/status/1649052544755470338

My assumption: aerodynamic forces were too great for the gimballing to fight against, therefore leading to a loss of control. EDIT: That, together with the assymetric thrust.
« Last Edit: 04/21/2023 02:44 pm by chopsticks »

Offline Herb Schaltegger

This forum software needs an upgrade to add a “laugh” reaction.
Ad astra per aspirin ...

Offline Danrar

  • Member
  • Posts: 55
  • Liked: 114
  • Likes Given: 60
SpaceX and Musk made a mistake on this concrete pad design. Everybody makes mistakes. They knew it could be a mistake.
They were expecting damage, they were not expecting this damage.
Part of me wished they had nailed it right because from an egoistical point of view, I will have to wait for a long time before seeing the next attempt.
But Musk has a ton of money and is hellbent on space. So in the long run, this mistake is NOT a problem. If anything it will teach the wealthiest man in the world some humility. I do not forget it's his stubbornness, dedication, wealth that made this rocket possible in the first place. Those some "qualities" will have drawbacks along the way. "Ne jetons pas le bébé avec l'eau du bain" as we say in French ;)
I just hope they yolo'd a little bit less the heat tiles design on the Ship. That's another area where I felt they were saying "screw heritage, we know better".
Considering the number of Ls he taken over the last 6 months to a year and yet he keeps carrying on I'm not sure he can be taught humility.

As for the heat tiles - that's not a particularly odd design or decision. The only real question is can that system support rapid turnarounds. From the basic standpoint of surviving a single reentry it should work no problem because functionally it isn't really that different from the Shuttle tiles.

Offline eeergo


I agree that we saw one of the HPUs explode shortly after liftoff, but now I'm not so sure that the other one failed.

Look at the end of Micheal Baylor's video. You can clearly see the centre engines gimballing while the other engines' plume remains straight.

https://twitter.com/nextspaceflight/status/1649052544755470338

My assumption: aerodynamic forces were too great for the gimballing to fight against, therefore leading to a loss of control.

To be clear, that's why I said "it was interpreted". However, Corvus appears to say no hydraulic failures occurred, which goes against the pretty evident first explosion.
-DaviD-

Offline Corvus Corax

  • Full Member
  • *
  • Posts: 176
  • New England
  • Liked: 310
  • Likes Given: 147
There was no stage separation or flip maneuver commanded.
The official SpaceX statement itself confirms this in that they state that the booster experienced multiple engine outs and then could no longer maintain attitude.
You may wish to zoom you screen in: the SpaceX statement says lost altitude, not attitude.

Flip start was commanded and expected (coinciding with commentators calling out start of separation, and camera switches timed to catch flip and separation, which then did not occur). Continued flipping was not intentional.

I had a conversation with a SpaceX engineer who corroborates this. The flip maneuver is apparently designed to be a pretty dramatic change in attitude that then reverses and swings the ship off the interstage just after MECO. HPU did not fail according to my source. TVC was operating fine during the flip maneuver, but he suspects that there was an anomaly with the software / interstage. Software got confused and the commands for the second half of the flip maneuver + MECO didn't execute as planned.

We literally saw a head-on shot of the HPU exploding, sending shrapnel flying everywhere, leaking a brightly-yellow-burning hydrocarbon into the exhaust stream for several seconds, and leaving a mighty hole in the stack where it was mounted.

A less clear event of the sorts happened shortly before "flipping" (either intentional or accidental), which many interpreted as the second, and last, SH HPU giving way too.

Regardless, the flip happened much earlier than BECO or separation was programmed for, both according to the published timeline and to live callouts. Even if the trajectory was off-nominal, both altitude and velocity were much lower than expected, so separation should if anything have been adjusted to happen later, at propellant depletion - not earlier. Furthermore, the separation couldn't have happened under booster thrust, no matter the attitude or angular rates. So in any case the first failure if we're to assume all the initial gyrations were intentional, was failure to effect BECO, not separation itself.

Are you sure no information was lost in your interview?

The source was quite clear. TVC was reportedly working fine up until the point where the flip swing was supposed to "throw" the ship. The initial off-axis maneuver was intentional, but the vehicle didn't know what to do with itself beyond a certain point due to an unknown anomaly. Source theorized that it was an edge-case in the flight telemetry parameters used to signal the second half of the maneuver. From that point, the vehicle started tumbling due to its off-program status. MECO never occurred because it was supposedly off-program.

Offline Danrar

  • Member
  • Posts: 55
  • Liked: 114
  • Likes Given: 60
They were literally one wayward concrete chunk from the entire orbital complex going up in a massive fireball. There's no way that you can look at what happened and not see that as a possible outcome.

Really?  Literally multiple millions of chunks of wayward concrete were blasted into the surrounding infrastructure,  but if there had been +1 more,   Disaster fireball ensues?  None of the millions of known chunks were wayward?  Just needed +1 more?
If one chuck lands in the wrong place inside the tank farm and causes methane to spew out yes absolutely one chunk.

Offline RoboGoofers

  • Full Member
  • ****
  • Posts: 1001
  • NJ
  • Liked: 868
  • Likes Given: 962
so lets list the problems (that we can know for certain from the videos alone) that need solving:
lots of engines out- a known problem cause there are new, better engines now.
exploding HPUs (presumably)- not a problem because they're switching to electric TVC.
stage sep- didn't work, maybe intentional, maybe something broke. needs solving.
lauch pad- a known problem cause there's parts for a flame trench and water deluge on site or on the way.

aside from the stage sep, there's nothing of value learned from these failures. For the other three, they can't be sure the new hardware will be any better. We'll see at the next test.

What, I assume, they do value is data on:
 propellant loading
 autogenous press
 propellant flows while thrusting
 structural stress and strain
 accoustics
 and lots of other stuff they can feed back into their engineering models.

though I suspect some of that data might be a bit dirty because of the other failures.

from my perspective as a spectator, this means we'll be spending a couple months at least watching some civil engineering, instead of watching more starship launches.

Offline Mark S

  • Senior Member
  • *****
  • Posts: 2284
  • Dallas, TX
  • Liked: 395
  • Likes Given: 80
I am also guessing that it will be much more than a year until the next test flight. Making a proper pad probably means bringing soil in, compacting it, making proper trenches, remaking the pad, the tower and installing the biggest rocket water deluge system known to earth, as well as proper GSE for rapid re-usability (in-situ manufacturing, enough electric power for it etc).

In the end, I think the pad is going to look less like today and more like this:

This is going to take some time, and I don't think the FAA would have let them fly in the meantime anyway, even if they wanted to, given the debris creation specifics of this flight.   


Much as I think that would be awesome, there is no room at Boca Chica for such a launch complex. There is barely room for what is already there, and that's after SpaceX expanded on the area they were originally given approval for. If you recall, Boca Chica was originally designed for Falcon 9 launches, possibly Falcon Heavy. Super Heavy and Star Ship only came along after the initial environmental planning and regulatory approvals.

They would have to expand south all the way to the Rio Grande to come close to that size of a launch complex. And that whole area is a tidal basin that floods at high tide and is a bare expanse of sand otherwise.

But, never say never. If Elon Musk were to make such a proposal to the Texas state authorities, I'm sure they would be happy to personally bulldoze the whole area for him. Environmental protection is not high on Austin's agenda.

Online novo2044

  • Full Member
  • *
  • Posts: 188
  • USA
  • Liked: 375
  • Likes Given: 45
They were literally one wayward concrete chunk from the entire orbital complex going up in a massive fireball. There's no way that you can look at what happened and not see that as a possible outcome.

Really?  Literally multiple millions of chunks of wayward concrete were blasted into the surrounding infrastructure,  but if there had been +1 more,   Disaster fireball ensues?  None of the millions of known chunks were wayward?  Just needed +1 more?
If one chuck lands in the wrong place inside the tank farm and causes methane to spew out yes absolutely one chunk.
The methane is in the horizontal tanks protected by the angled berm.  Could something have shot straight up and landed on it?  Sure in fact a ton of gravel probably did.  But take a look from a low angle and the methane portion of the farm is pretty well protected.

Quote
But, never say never. If Elon Musk were to make such a proposal to the Texas state authorities, I'm sure they would be happy to personally bulldoze the whole area for him. Environmental protection is not high on Austin's agenda.
If Texas/Federal government passed a law that said: SpaceX can do as they please, there would be a ton of craters around Boca Chica, a cemetery for the honored dead, and a Starship preparing for a Mars today I suspect.

Offline cuddihy

  • Full Member
  • ****
  • Posts: 1251
  • Liked: 580
  • Likes Given: 935
I was expecting a better flight.  At least to staging.   Didn't expect so many engines out.   Thought they would have done their engineering better on the launch pad.

I don't understand how the engineers at SpaceX didn't anticipate that they would need to make sure the energy from a Booster twice as power as the Saturn-V was properly diverted in a way that wouldn't damage the launch pad and surrounding area.

Between this and Boring Company’s awful plan to make aggregates with in-situ materials, Elon needs to hire or befriend some better civil engineers.

I've waited for several days for the air to clear and more info to become available, but it's time say something.

Frankly, Elon had good people helping him do this for many years. They successfully built him west coast and east coast launchpads. He decided they weren't moving fast enough / were being too "traditional" for Starship and let them go two years ago. I know one very senior engineer manager for him who was pushing for a more traditional flame trench/divertor at BC who Elon got tired of hearing from and fired. This is the result...this one's on Elon, personally, IMHO. People in SpaceX repeatedly warned him the risks of damage from the concrete. The tweet several months ago was his belated acknowledgement that they were probably right, but it was too late at that point, he was committed to the current flat pad at that point.

Offline sferrin

  • Full Member
  • ****
  • Posts: 722
  • Utah
  • Liked: 909
  • Likes Given: 760
I agree with Ed. Ground testing for full duration at full power for a vehicle this large and complex is a must. I doubt even Elon can continue to lose SH after SH with major OLM repairs after every attempt by not doing that.
You're conflating things.  A rocket is never going to sit on its launch pad for a full-duration burn.  SLS never did that.  Saturn V never did that.  Falcon Heavy never did that.
"DARPA Hard"  It ain't what it use to be.

Offline lawlessl

  • Full Member
  • *
  • Posts: 119
  • UK
  • Liked: 125
  • Likes Given: 99
The first step to learning from a failure is to admit that there was a failure.

Was there a failure? There was certainly no lack of success, and I doubt any neglect or omission either.

To me it was the most wondrous proof of engineering and testing. Forget can crushers. This was acrobatics under extreme load. I was amazed that it held together so well they had to destroy it with explosives. Imagine if someone had suggested doing tumbles at 35km as a testing strategy. Never mind wild claims that this situation was predicable. The reality is that no one could have predicted its structural success.

Now what was the cause. How about assuming the concrete was doing fine for the initial long hold down. To my eyes I see no problems. I assume the sequence was to go to lift off power and release immediately. If it was to fail, then let it be moving away from the costly part. The concrete surface held, but the ground below did not. It compressed, creating a  concrete inverted dome, which shattered all at once The thrust got beneath the concrete and ripped it up.

The debris would be like water in a pelton wheel. Some of the lumps of concrete would hit the OLT legs and bounce up into the engines. The next few minutes are a consequence of that single event.

What I would like to see would be a composite picture of the engines to see how many had damage to the nozzles. Across multiple frames it might be possible to detect. I would also like to see the engine damage in relation to the OLT legs. There seems a definite pattern. The shape of the hole under the OLT is also of interest.
« Last Edit: 04/21/2023 03:09 pm by lawlessl »

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0