totm nov 2023 SpaceX Discussion Thread

[Exoscientist]

15 hours ago, Ultimate Steve said:

@Exoscientist What would convince you that Raptor has become reliable?

We don't know exactly how many engines lit for the landing burn as the indicator didn't work, and who knows, there might have been an engine out there, though I doubt it. There were 2 engine outs, which were compensated for, across 33+10+13+6+2? attempted engine starts, a score of 62/64 plus or minus one, 96.8% reliable on this flight. One of those failures was in a particularly intense environment (landing burn, that thing was going FAST) on the second ever attempt, the first of which was, to my knowledge, foiled by a clogged filter.

While that's obviously not amazing, I don't think it is worth doomering about on flight 4 (IIRC flight 3 of Raptor 2).

While I don't think it is an amazing comparison, Falcon 9 was 39/40 cumulative through Flight 4, and the J-2 (at least on the Saturn V, I didn't look up Saturn I, or the Merlins on Falcon 1 for that matter) was 23/26 by flight 4 of the Saturn V. I'd have to go back and do a lot more counting for a cumulative Starship number though.

 

 I was watching a stream without the SpaceX commentary. Was there a call out that at a certain time during reentry there was supposed to be an engine firing to slow down the ship? Obviously, the degree of damage to one of the flaps during reentry was not normal, which suggests there should have been a burn then.

 There is uncertainty if there was a landing burn for the ship just before touchdown. The graphic at the bottom right showing which engines are lit did not come on. On the other hand the ship did reorient itself just before landing. It is possible the RCS and the flap control surfaces are sufficient to reorient the ship when tanks are near empty.

 A major irritation of mine is SpaceX is continuing to promote the idea the Raptor is already at sufficient reliability level for an engine intended to carry crew. The reality is it is nowhere near that level which should be 99.9+%

  Bob Clark

Edited June 7 by Exoscientist

[Ultimate Steve]

18 minutes ago, Exoscientist said:

 

 I was watching a stream without the SpaceX commentary. Was there a call out that at a certain time during reentry there was supposed to be an engine firing to slow down the ship? Obviously, the degree of damage to one of the flaps during reentry was not normal, which suggests there should have been a burn then.

 There is uncertainty if there was a landing burn for the ship just before touchdown. The graphic at the bottom right showing which engines are lit did not come on. On the other hand the ship did reorient itself just before landing. It is possible the RCS and the flap control surfaces are sufficient to reorient the ship when tanks are near empty.

 A major irritation of mine is SpaceX is continuing to promote the idea the Raptor is already at sufficient reliability level for an engine intended to carry crew. The reality is it is nowhere near that level which should be 99.9+%

  Bob Clark

There is not and has never been a plan for a ship entry burn unless I've missed something pretty major. There's also currently no entry burn for Super Heavy, unlike Falcon 9. The flap hinges were a known problematic area before the flight (there is a preview clip of the EDA interview showing Elon predicting that if it failed, that would be the failure mode). They have had a redesign in the works for Starship version 2 for a while now.

Landing burn happened for sure. Video showed illumination at the same time the engines were supposed to turn on (unclear what the earlier illumination was so could be a fluke I guess). There were callouts for landing burn startup, landing burn in progress, and landing burn shutdown. Telemetry showed the flip, ship slowing down during flip, coming to a standstill, and then falling over as if it was in the water. Video was hard to make out but appeared to show a violent impact at about the same time the telemetry indicated it had fallen over after landing in the water.

For it to had not have done a landing burn, the lighting would have to be coincidental, the callouts would have had to be lying, the telemetry would have had to be lying, and there would need to be an alternate explanation for what the video showed. Not impossible but on the edge of impossible. Engine telemetry to the stream did indeed fail.

Unless you're specifically referring to the reliability needed for crewed propulsive landings, which would be a different category, I will note that SSME and Merlin are also not at your threshold for crew rated reliability. You're talking 1 failure in 1000 firings. Maybe the RD-107/108 meet that if you measure in a certain timeframe? There are a number of 100% reliable engines, of course, at the sample sizes they have been fired at (nowhere near 1000).

[Flying dutchman]

Has anyone calculated where starship splashed down? Perhaps an earth imaging satellite has picked it up.

[cubinator]

6 hours ago, Exoscientist said:

There is uncertainty if there was a landing burn for the ship just before touchdown. The graphic at the bottom right showing which engines are lit did not come on. On the other hand the ship did reorient itself just before landing. It is possible the RCS and the flap control surfaces are sufficient to reorient the ship when tanks are near empty.

I'm extremely confident the ship performed a landing burn, because:

-We see a light flash on from below as the ship reorients

-The velocity indicator gradually slows to near zero, then increases, characteristic of the ship hovering above the water and then allowing itself to sink in

[AckSed]

Here's a thing.

With that footage of the hot atmospheric gas flaring through the hinges, the odd diagonal extensions of the heatshield on the leeward side made sense. They were expecting this to happen, and assumed that the tiles covering the inside of the flap and hinge were enough.

[Exoscientist]

20 hours ago, cubinator said:

I'm extremely confident the ship performed a landing burn, because:

-We see a light flash on from below as the ship reorients

-The velocity indicator gradually slows to near zero, then increases, characteristic of the ship hovering above the water and then allowing itself to sink in

 There still is the issue the engine lit indicators on the SpaceX video stream did not light up for the ship landing burn. Note that for both this flight and previous flight they did correctly light up for landing burns.

EDIT: the engine lit indicators worked correctly both times showing which engines fired during the landing burn for the booster.

 

  Bob Clark

Edited June 8 by Exoscientist
Clarified which engine lit indicators worked during landing.

[Codraroll]

15 minutes ago, Exoscientist said:

There still is the issue the engine lit indicators on the SpaceX video stream did not light up for the ship landing burn. Note that for both this flight and previous flight they did correctly light up for landing burns.

The footage still showed a splashdown at very moderate speeds, judging by the movement in the video. If there had been no landing burn, it would have been a much more sudden impact with the sea. Not to mention that all other telemetry showed indications of a landing burn, like speed and orientation.

Face it, you're stuck with a bogus conclusion and you're grasping at straws to justify it instead of changing it according to the observed evidence.

[RCgothic]

Starship: *Fin fails during first true test of re-entry.*

"This was clearly Raptor's fault."

[sevenperforce]

7 hours ago, Exoscientist said:

 I was watching a stream without the SpaceX commentary. Was there a call out that at a certain time during reentry there was supposed to be an engine firing to slow down the ship? Obviously, the degree of damage to one of the flaps during reentry was not normal, which suggests there should have been a burn then.

No, there was not and has never been any plan for a re-entry burn for the ship. SpaceX uses supersonic/hypersonic retropropulsion for the Falcon 9 booster during its re-entry, but there has never been any possibility of some sort of ultrasonic retropropulsion for entries from orbital velocity. The ship is not in the correct orientation for doing that and it would not make any sense.

"which suggests" is doing a lot of heavy lifting there. It's a non sequitur. I might as well say "Apollo 12 getting struck by lightning is not normal, which suggests it should have deployed parachutes during ascent."

7 hours ago, Exoscientist said:

 There is uncertainty if there was a landing burn for the ship just before touchdown.

There is not.

7 hours ago, Exoscientist said:

The graphic at the bottom right showing which engines are lit did not come on. On the other hand the ship did reorient itself just before landing. It is possible the RCS and the flap control surfaces are sufficient to reorient the ship when tanks are near empty.

It is not.

48 minutes ago, Exoscientist said:

 There still is the issue the engine lit indicators on the SpaceX video stream did not light up for the ship landing burn. Note that for both this flight and previous flight they did correctly light up for landing burns.

There were no other ship landing burns on this flight or on the previous flights.

Even if RCS and the flap control surfaces were sufficient to reorient the ship, this would not explain (a) why we saw a landing burn, (b) why we had landing burn callouts, or (c) how the ship came to a hover over the water. What's more likely -- that the laws of physics were suspended to allow a plummeting object to suddenly freeze in place in midair, or that the graphic didn't update properly?

7 hours ago, Exoscientist said:

 A major irritation of mine is SpaceX is continuing to promote the idea the Raptor is already at sufficient reliability level for an engine intended to carry crew. The reality is it is nowhere near that level which should be 99.9+%

How is SpaceX "continuing to promote" this idea? Simply by having a contract? If having a contract to deliver crew means that you are promoting the idea that any engines you would use to carry that crew are already at full reliability, then that is a problem for a number of NASA contractors who have contracts but don't even have engines yet.

9 hours ago, Ultimate Steve said:

I'm beginning to see where Exoscientist is coming from. I would still argrue that in the context of the Starship program, the current reliability is within the realm of what is expected at this stage. But with conventional thinking, for something that is to power a moon lander, even 98.2%, let alone 94.1%, or 74%, is an absurdly low number.

Agreed that current reliability is where we would expect it at this stage.

Additionally, as long as we don't have catastrophic or sympathetic failure modes, engine-out capability should be treated a little differently.

[PakledHostage]

8 hours ago, Exoscientist said:

 Is it possible the high temperature alloys used for the engine combustion are being confused with the metals used for the tanks?

  Bob Clark

I was thinking the same thing. I had never heard of either SX300 Inconel or SX300 stainless steel.  I assume that they are both proprietary SpaceX alloys, judging by the "SX" prefix. According to the Wikipedia article where SX300 Inconel is mentioned,  that alloy is used in SpaceX engines. I guess it is possible that they have a SX300 stainless alloy that shares a name with their Inconel alloy... despite the confusing nomenclature,  it would make some sense under these circumstances. 

[darthgently]

4 minutes ago, PakledHostage said:

I was thinking the same thing. I had never heard of either SX300 Inconel or SX300 stainless steel.  I assume that they are both proprietary SpaceX alloys, judging by the "SX" prefix. According to the Wikipedia article where SX300 Inconel is mentioned,  that alloy is used in SpaceX engines. I guess it is possible that they have a SX300 stainless alloy that shares a name with their Inconel alloy... despite the confusing nomenclature,  it would make some sense under these circumstances. 

I agree, this seems like confusion on Wikipedia's part.  I don't see an Inconel rocket body being any more economical than carbon fiber.  I also don't think Inconel would dissolve as easily as in the vid of the flap, but just guessing

[JoeSchmuckatelli]

1 hour ago, sevenperforce said:

@Exoscientistsaid:

 A major irritation of mine is SpaceX is continuing to promote the idea the Raptor is already at sufficient reliability level for an engine intended to carry crew. The reality is it is nowhere near that level which should be 99.9+%

 

... 

current reliability is where we would expect it at this stage

Aren't we getting into circular argument territory, again? 

We can agree that another company pursuing a NASA contract in this or past times would test and test and test on the ground - quietly (and without observation by the space enthusiast community) experiencing failure after failure until they did not - and only then progress to a flight (much less crewed flight)...

Doing so would make their NASA - flight reliability figures look respectable. Correct? 

But this company did not start this program in pursuit of nor in accordance with a NASA contract (AFAIK): BFR / Starship & Booster began as an internal program / vanity project with only the company itself as a client.* 

Only later did NASA say 'interesting stuff yer doin over there, Tex, how's about you pitch us on doing our thing with your rocket?'

So to come in at this point and say  "because NASA has interest" that this project must align with some other arbitrary SOP attributable to different programs is inapt. 

Our cribbing about reliability figures for this very public and highly entertaining iterative prototyping project and comparing it to the 'completed item under contract' reliability figures is like asking why fish don't have stems**

 

 

 

 

 

* (As such - it does not matter what other people did before; they're free to ground test or test fly at will).  Precedent is meaningless when the fact scenario does not align... just as "industry standards" are not applicable to the prototype / vanity project category of work. 

 

** - because apples do, and they float in water, and they're delicious, thus because fish are found in water and are delicious, they should have stems, too, n'est cie pas? 

Edited June 7 by JoeSchmuckatelli

[Entropian]

13 hours ago, mikegarrison said:

Maybe Musk just doesn't like the term "super-alloy" (or "superalloy").

The rocket is called "Superheavy."  I highly doubt he has a problem prefixing the word "super" to things

[JoeSchmuckatelli]

9 hours ago, Exoscientist said:

uncertainty if there was a landing burn for the ship just before touchdown

Look at the acceleration there at the end of the flight 

It's possible the telemetry data wasn't being picked up by the stream.  It was pretty obvious from watching that the video link and telemetry data displayed were separate data streams.  Several times during the flight, video would cut out but the telemetry data kept changing.  However at the end, we had video but no telemetry. 

 

Shrug. 

[CBase]

10 hours ago, Exoscientist said:

which should be 99.9+%

Starship and Raptor to convention rockets is more like cloud computing to mainframes.

One gains reliability by avoiding failures, the other compensates individual failures by controlling software stack. But both can support highly critical operations.

The center 3 engines are most critical part during final landing and do only allow for single failures. Which means you need 96.8% single reliability to have 99.9% against double failure. Is raptor currently up to this ? propably not. But again since the produce more than the 3 critical ones, they might be able to select best performing on final assembly control. So again just by collecting data, correlate to test performance they could get to a point that is acceptable.

[.50calBMG]

I would like to see the actual reliability for the actual number of the current raptors that have failed mid flight vs the ones that succeeded, because I bet the reliability is in the ballpark of most engines out there. The closest engine I can think to compare it to complexity wise would be the RS-25s, which may I remind some people still had a test stand premature shutdown on the sls green run tests after flying on the shuttle for how many years? No engine will ever be perfect, but I bet raptor is doing about as well as any other engine at this stage of development.

(I feel like I've said this before...)

3 minutes ago, .50calBMG said:

 

I would also like to point out that the failures during the boostback and landing burns from IFT-3 were fuel starvation, not the engines exploding.

Edited June 7 by .50calBMG
Don't know what my phone is doing, somebody help

[mikegarrison]

2 hours ago, PakledHostage said:

I was thinking the same thing. I had never heard of either SX300 Inconel or SX300 stainless steel.  I assume that they are both proprietary SpaceX alloys, judging by the "SX" prefix. According to the Wikipedia article where SX300 Inconel is mentioned,  that alloy is used in SpaceX engines. I guess it is possible that they have a SX300 stainless alloy that shares a name with their Inconel alloy... despite the confusing nomenclature,  it would make some sense under these circumstances. 

Or Musk just screwed up in his tweet.

I am pretty sure the rocket body is not made up of a nickel super-alloy, because they are hard to work with. And also expensive. But that's all the info I was able to find on SX300.

That said, Inconel 718 is supposed to have "good weldability", and the Wikipedia claim is that SX300 is mono-crystal Inconel 718.

[Flavio hc16]

Hearing from Musk interview with "Ellie in space" at around the 2.20 minutes mark, Elon said that the superheavy or the ship( it's not very clear to me) missed the targhet landing zone by 6 kms, but both landed correctly.

 

https://youtu.be/tjAWYytTKco?si=H8S1E7y9crEzc8a7

 

 

[Minmus Taster]

16 minutes ago, Flavio hc16 said:

Hearing from Musk interview with "Ellie in space" at around the 2.20 minutes mark, Elon said that the superheavy or the ship( it's not very clear to me) missed the targhet landing zone by 6 kms, but both landed correctly.

 

https://youtu.be/tjAWYytTKco?si=H8S1E7y9crEzc8a7

 

 

Probably the booster, ship didn't really have a landing zone beyond "Please try and land in this ocean and we'll worry about the details later". Probably no catch on the next flight then. 

[PakledHostage]

1 hour ago, Minmus Taster said:

Probably no catch on the next flight then. 

Those grid fins did seem to be working really hard yesterday morning.  I briefly tensed up thinking it was going to lose control again. It recovered, but I wonder how close it was to failure? You'd think they'll want to be able to stay more in the middle of the "controlabilty corridor" before they try a catch?

[JoeSchmuckatelli]

2 hours ago, Flavio hc16 said:

the ship( it's not very clear to me) missed the targhet landing zone by 6 kms

 

2 hours ago, Minmus Taster said:

booster

I watched the video - Booster went to a "precise location" and it was Ship that was "technically" 6km off geographically. Start with 2:00

Booster going to 'precise location' and 'zero velocity landing' supports chopstick attempt. 3:40

For those in the 'steel vs alloy' debate he keeps referring to the SX300 as steel. 

Edited June 7 by JoeSchmuckatelli

[Flying dutchman]

15 hours ago, Flavio hc16 said:

Hearing from Musk interview with "Ellie in space" at around the 2.20 minutes mark, Elon said that the superheavy or the ship( it's not very clear to me) missed the targhet landing zone by 6 kms, but both landed correctly.

 

https://youtu.be/tjAWYytTKco?si=H8S1E7y9crEzc8a7

 

 

I'm just going to say it: elon sometimes sounds like a broken record. Every interview he tells the exact same story about reusability.

 

I agree with the story, but i'd like to hear some more in  depth stuff.

Edited June 8 by Flying dutchman
Forgot the "h" in "the"

[mikegarrison]

3 hours ago, Minmus Taster said:

Probably the booster, ship didn't really have a landing zone beyond "Please try and land in this ocean and we'll worry about the details later". Probably no catch on the next flight then. 

I am sure they had a landing zone, defined in their test plan.

1 hour ago, JoeSchmuckatelli said:

For those in the 'steel vs alloy' debate he keeps referring to the SX300 as steel. 

I am not convinced that Musk knows this kind of detail. For whatever that is worth. I've watched some of the video interviews with him, and he talks a lot more like a manager than an engineer.

But it's probably not worth getting into the whole "Musk is not an engineer" thing again.

[DAL59]

https://www.nasa.gov/news-release/nasa-exploring-alternative-mars-sample-return-methods/

Starship proposed for Mars Sample Return!

[JoeSchmuckatelli]

22 minutes ago, DAL59 said:

https://www.nasa.gov/news-release/nasa-exploring-alternative-mars-sample-return-methods/

Starship proposed for Mars Sample Return!

Cart? 

Meet Horse! 

Horse? 

Push! 

- not that I don't recognize the promise... But they're going to have to crash 3 into the moon (at least) before landing there, then 3 into Mars before landing (hopefully AFTER deployment of several Starlink constellations) and then figure out ISRU and... 

... 

Big dreams.  2040?  Might buy 2035 if we see Moonshot by 2028.