Talk:SpaceX Starship

Static Fire Tests

I've noticed that I can't find anything on the subject of Starship (SN) static fire tests or the implosions/explosions caused by them. I'd have thought that these were quite important pieces of information to include? It could just be that there aren't enough reliable sources to cite; I haven't checked, but I thought I'd take note of it here. — Preceding unsigned comment added by Vgj843df (talk • contribs) 10:35, 3 August 2020 (UTC)

Look in the other article, Vgj843df, Starship development history. That sort of detail does not belong in this article, given the other article, which exists as a result of a long Talk page discussion a year or so ago. Cheers. N2e (talk) 04:27, 8 September 2020 (UTC)

Is history getting too detailed?

The history section describing 2020 progress is getting fairly granular, which I am not opposed to in the abstract, except that seeing as the page Starship development history exists, that sort of thing might best be reserved for there. Whether we're at a point where the summary on this page should be cut down is a ymmv thing, I guess. Thoughts? BlackholeWA (talk) 05:37, 5 September 2020 (UTC)

Yes, agreed, BlackholeWA. We had a previous Discussion that led to the creation of that other article, Starship development history; but a lot of fairly detailed history stuff on testing is being added to this article (SpaceX Starship) by various editors, and it appears no editor has been catching it, reverting it, and suggesting to the editor adding it that it belongs in another article. Now, will take some editor with a bit of time willing to cull the extraordinary detail from this article, and move the appropriate bits to the development history article.

But do feel free to either do that; or flag the excessive stuff with article improvement tags. Cheers N2e (talk) 04:25, 8 September 2020 (UTC)

Payload to LEO

Image says 150 tons. Description on right says 100 tons. Which is right? Silenceisgod (talk) 03:31, 24 October 2020 (UTC)

100 tons is the closer to test correct number based on 3 SL Raptors with 2.2MN of thrust Dvdcrr77 (talk) 10:50, 10 December 2020 (UTC)

A Commons file used on this page or its Wikidata item has been nominated for deletion

The following Wikimedia Commons file used on this page or its Wikidata item has been nominated for deletion:

• Starship SN8.jpg

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 10:24, 3 December 2020 (UTC)

SN8 failure?

isnìt better to say "partial success"? it is only the last part a failure, and the test was about all. a failure would be an explosion on the launch pad. --Dwalin (talk) 23:10, 9 December 2020 (UTC)

SN8, not SN9. We should describe it however WP:RS describe it. It's still very early, and I'm sure more will come in, but here's one example: [1]. SpaceX Starship SN8 explodes after first successful high-altitude flight. "successful". And looking at others, I see a lot using the word "successful", but also a lot using the word "exploded on landing". I don't see much of anything describing it as a failure yet.

If I had to call it right now, I don't think that this table will end up calling it an unequivocal failure nor an unequivocal success in the end. Right now I'd say "Partial success" is best supported. Leijurv (talk)

Currently on Wikipedia for spaceflight and rocketry related lists, this is described as a failure because it was a destruction of the launch vehicle during the primary mission. This is why Falcon 9 landing failures aren't classified as failures because landing is a secondary objective or why 2nd stage fragmentation in orbit aren't classified as launch failures, because they occurred after completing the primary objective. Launches like Astra's Rocket 3 are classified as failures as the vehicle was destroyed during the primary mission, even though it was a high risk R&D flight. Since the landing of SN8 was a necessary part of the flight and was not a secondary objective, this test is labeled as a failure. (partial failures are missions where an anomaly in the primary mission occurs, but is still able to meet all objectives (like Atlas V L-no.10), so SN8 doesn't qualify). --Jrcraft Yt (talk) 01:07, 10 December 2020 (UTC)

(humorous response) What makes you say that the launch vehicle destruction happened during the primary mission? I don't think anyone can disagree that it was the impact with the ground that immediately caused the explosion, no? So it landed in one piece, THEN exploded.

(actual response) Thanks for sharing that, I didn't know. I think we might need to think about what the mission goal was here then? You are assuming that landing was a necessary part of the test, and because it didn't land, today was a failure (i.e. it did not succeed at its mission goal). I don't think that's true. From everything I've seen, the goal was always to collect data without an overriding intent or goal to land. In other words, I don't agree that landing was the primary mission objective. E.g. see this (imagine citing a meme tho). Leijurv (talk) 02:41, 10 December 2020 (UTC)

The Dragon In-Flight Abort Test is listed as a success, although the launch vehicle was obviously lost. The survival of the Falcon 9 was not critical to the test. So the test was a success despite the loss of a vehicle. Fcrary (talk) 04:40, 10 December 2020 (UTC)

Right now, it's listed as a partial failure, I say keep it like that. It shouldn't be marked as a success since the starship program revolves on being able to recover both stages. Test flight or not, it should be classed accordingly. We don't give test flights any special success/fail criteria on any other page, so we shouldn't here. --Bvbv13 (talk) 02:56, 10 December 2020 (UTC)

I disagree. The purpose of the test was to obtain flight data and learn how to improve future (preliminary, test) versions of Starship. SpaceX never said a successful landing was necessary for a successful test. In fact, they said exactly the opposite. We should use the same criteria for success as the criteria the operator announced in advance of the test. We did that for the Schiaparelli EDM Mars lander. It crashed, but the success criteria announced in advance said a landing was not required. Success simply required getting enough data to determine how the EDL system worked, and, if it didn't, why it didn't. Fcrary (talk) 04:40, 10 December 2020 (UTC)

We've catalogued rocket launches equally the for every rocket. We either have successful launches, partially successful launches or launch failures. There is also the "other outcome" data point. Unless we want to go back to ALL entries that contain launch stats, it stays like this. Every time a failure occurs, data is collected. That doesn't constitute revising launch catalogues for all of Wikipedia. We don't need separate "collected data before failure" and "didn't collect data before failure" tags. There are going to be launch failures and they need to be categorized as such instead of trying to find a way to claim 100% success. Yes, the point was to collect data and that part was successful, that doesn't mean it didn't explode upon landing failure. Recovering the vehicle is an integral part of the starship system. Same reason why the first launch of Astra's rocket 3 isn't classified as a success. Yes, the point was to get data, but the rocket still failed. There is no reason to make starship launches any different from all previous ones. There are going to be some non-green boxes in the chart and that's ok, that's how this kind of rapid prototyping works. (We didn't do that for the Schiaparelli EDM lander, in charts , it's listed as spacecraft failure). --Jrcraft Yt (talk) 05:11, 10 December 2020 (UTC)

Yes, the point was to collect data and that part was successful Okay good. Then: This is why Falcon 9 landing failures aren't classified as failures because landing is a secondary objective or why 2nd stage fragmentation in orbit aren't classified as launch failures, because they occurred after completing the primary objective. The rocket exploded after completing the primary objective. This is exactly like a Falcon 9 failing to land in that case. Leijurv (talk) 08:14, 10 December 2020 (UTC)

This isnt orbital launch not even a suborbital launch. As it was lower than some amateur rockets. Whats more important this test was more of proof of concept, that you can do it, that you can control rocket on its side, that you can do belly flop. It was never done before. We should not mistake program goal with test goal. Even if program goal is to land from orbit or suborbital space flight, this test was none of them. In fact there is Elons tweet where he writes test goals: "Good Starship SN8 static fire! Aiming for first 15km / ~50k ft altitude flight next week. Goals are to test 3 engine ascent, body flaps, transition from main to header tanks & landing flip."), by it its full blows success. Altitude was later changed to 12.5km. There is nothing said about successful landing. Janncis (talk) 08:56, 10 December 2020 (UTC)

yes. a SN20 that ends like that will be a complete failure. a SN8 is, it you don't want to say a success, a partial success. --Dwalin (talk) 10:55, 10 December 2020 (UTC)--Dwalin (talk) 10:55, 10 December 2020 (UTC)

A new way to present the outcome of testing instead of success or failure

I see the other previous discussion on whether the test is a failure or success. The problem is a test can be classed as a success by just getting some data but failing to achieve anything. I propose instead of labelling the prototype tests as a success or a failure, it is instead labelled as "All test objectives achieved,' 'Most test objectives achieved,' 'some test objectives achieved' and 'failure.' Green would be for all test objectives achieved, red for failure, yellow fo most test objectives achieved and light red for some test objectives achieved. Input on changes to wording is welcome of course -AndrewRG10 (talk) 02:34, 10 December 2020 (UTC)

I agree in spirit that the breakdown of "primary mission objective achieved or not" just does not work well with what SpaceX is doing here. When they do a launch into orbit, it's clearly applicable, such as the examples above of F9. For a test that is explicitly for data gathering (stated beforehand), a test that was explicitly stated as having a 1/3rd chance of completing in full ([2])? I don't think we can say what the primary mission objective is. If it's data gathering, SpaceX has said very clearly that that was successful ([3] [4] [5] [6]). I think we would need an alternate reliable source to support the idea that the primary mission objective was to land, and not to collect data. Leijurv (talk) 02:51, 10 December 2020 (UTC)

Right now, it's listed as a partial failure, I say keep it like that. It shouldn't be marked as a success since the starship program revolves on being able to recover both stages. Test flight or not, it should be classed accordingly. We don't give test flights any special success/fail criteria on any other page, so we shouldn't here. --Bvbv13 (talk) 03:00, 10 December 2020 (UTC)

I didn't say it should be listed a success. I said we should completely change how the test flights are categorised instead of success failure or partial success, it should be listed as a failure, all test objectives achieved or partial test objectives achieved. Obviously, SN8 would go into partial objectives achieved. -AndrewRG10 (talk) 04:43, 10 December 2020 (UTC)

I disagree. What if the goal of the program is to build, test and crash vehicles until SpaceX learns enough to build one which can land safely and reliably? Landing is the goal of the program not the goal of any single test within that program. If a particular test's goal is to gain information for future use, and it produces that information, then the test is a success by any reasonable standard. Even if it only gained the large majority of the desired results, it would still be considered a success. Note that the Galileo mission is considered a success, despite the fact that the spacecraft's high gain antenna problem meant that it only accomplished 80-90% of its planned objectives (as estimated by NASA.) We don't need to list "partial success" for activity which falls short of 100% Fcrary (talk) 04:51, 10 December 2020 (UTC)

That breaks how we've cataloged all rocket flights on Wikipedia. When we have a launch that doesn't meet all requirements, it goes under partial failure/partial success. Unless we want to change every rocket entry on Wikipedia, it stays like this. We either have successful launches, partially successful launches or launch failures. There is also the "other outcome" data point. Test flight or not, they need to be catalogued the same way. It's the same reason why the first launch of Astra's rocket 3 isn't classified as a success. Yes, the point was to get data, but the rocket still failed. All entries in Wikipedia don't need to change because a Starship prototype crashed into the ground. Creating a disparity in this article from the accepted norm so that we don't have to see something other then a green success box for a launch outcome is absurd. --Jrcraft Yt (talk) 05:11, 10 December 2020 (UTC)

Actually, that is not how we categorize all rocket flights. For the Falcon 9, we list launch success/failure and landing success/failure separately. From what I've seen on the Electron talk page, that's also the plan for Electron. Perhaps we should describe the Starship SN8 flight as a successful mission but a landing failure. Fcrary (talk) 06:03, 10 December 2020 (UTC)

That is a good point, we could do that, one reservation I have though is that for F9/H or Electron, recovery isn't mission critical. Starship is another story. It is mission critical, like shuttle. When it's carrying people back to Earth or to Mars, landing is mission critical. It's up to the talk page to decide. Personally, I believe that the landing of Starship is mission critical since it's designed to land humans on the Moon, Earth and Mars aswell as be quickly refueled and relaunched. So it should be factored into overall mission success. Whether or not the Space Shuttle Orbiter returned was directly connected to mission success so I don't think Starship should be any different. --Jrcraft Yt (talk) 06:18, 10 December 2020 (UTC)

We should decide on the talk page, but I think you are mixing up mission critical for the final, operational Starship, and mission critical for a test of the SN8 vehicle. SN8 is nothing like what an operational Starship would be like. So the mission success criteria for its test flight should not be confused with what an operational Starship would need to do. If they hadn't put landing gear on SN8 at all and just dumped it into the ocean, rather than attempting to land it (which was speculated about before today), would you call the test a mission failure? Because they didn't do something they never intended to do with this prototype? For the Starship program, eventually developing safe and reliable landing systems is a mission goal. But for the SN8 test, simply getting data which contributes to that goal is a mission success. Fcrary (talk) 06:50, 10 December 2020 (UTC)

The launch outcome ended and explosion during the failed landing. Just like we don't characterize the Schiaparelli EDM lander (even though it's primary purpose was to collect data for future probes) as a success. Just because it's a test flight, we shouldn't claim success simply for getting data. Every launch gathers data. The fact is that this was a mostly successful test where most of the mission went as planned. Only one part went wrong and that was the landing burn. Hence why it's classified as partial failure, because only one part failed. You can't claim 100% success on a mission that had aspects fail on the vehicle leading to a total destruction of the rocket (no matter how high the risk was). At most, you could change it to "mostly successful". But not full success. Even on risky test flights, there hasn't been an instance where we've categorized a total vehicle destruction as fully successful. With that same logic, Astra's rocket 3.2, Virgin orbits demo flight, Electron's first flight and more were all 100% successful because they prioritized data collection over payload. --Jrcraft Yt (talk) 07:02, 10 December 2020 (UTC)

I think you are getting fixated on the idea that "success" means "100% success". For some things, like launching a satellite into the intended orbit, that can make sense. But for any sort of test or experiment, there is no such thing as 100% success. If 100% is the criteria, then no one has ever launched a successful scientific spacecraft. Hubble wasn't (due to the initial problems with its optics). Kepler wasn't; pointing stability wasn't quite as good as planned, and they needed an extended mission to achieve their prime mission goals. Even the most wildly successful ones haven't managed better than 99%. So I think we should just drop the whole idea of 100% success being required to say a mission or a launch was "successful." 90% is (as in the case of Galileo) close enough to 100% to call it a success.

And you didn't comment on the criteria for success. Look, the Starship SN8 did not successfully test landing, but it also did not test other things the final, operational Starship will need. It did not test life support systems. It did not test reentry or heat shields. It did not test in space refueling. It did not because the SN8 test was not supposed to do so. SN8 was purely intended to collect data for future iterations of the design. If we judge a mission or a test by its intended goals, as I think we should, SN8 was a success. The got data on all the things they were interested in, and all but one part of the test worked. The landing didn't, but since the goal was to collect data on whether or not the system worked, the test accomplished its mission. Fcrary (talk) 08:20, 10 December 2020 (UTC)

Will comment here to, there is Elons tweet [1] about tragets, if we go by them, its full success as altitude was later changed. Janncis (talk) 09:04, 10 December 2020 (UTC)

Disappointing that I have to restate my proposal because people turned this into a what is a success discussion, that's for the talk section above. I propose that in the testing section, (not anywhere else on Wikipedia or in the real launch section) that success, partial and failure are removed and replaced with 'all test objectives achieved,' 'most test objectives achieved' and 'failure.' It would look like this.

This system is designed to make it impossible to have an argument over what category the test goes under because it's pretty clear if it didn't get all of the test goals complete, ie, SN8. -AndrewRG10 (talk) 10:27, 10 December 2020 (UTC)

Sounds great, lest do it. SN8 flight then should be All test objectives achieved as per Elon's tweet, But I will suggest in that case adding an other choice of "Test objectives unknown" since no one here will publish test checklist. All we have to go by is Elons tweets and words.Janncis (talk) 18:03, 10 December 2020 (UTC)

Can we split the thrust level for Second Stage sea/vac

Just for clarity.

Dvdcrr77 (talk) 10:47, 10 December 2020 (UTC)

launch outcomes

is it correct to put in the general page the outcomes of the prototype? spacex is not NASA. there will be at least 50 launch before first payload. (if i don't remember whrong elon). NASA is "good the first". --Dwalin (talk) 17:36, 10 December 2020 (UTC)

I like it. My only concern is using the word "all". I don't think any test is ever 100% successful, and I'd hate to see arguments about a 99% success not accomplishing "all" of the goals. But I can't think of a better way to phrase it. Fcrary (talk) 19:34, 10 December 2020 (UTC)

if falcon 1 was a "normal" rocket. starship is totally another kind of rocket. they uses lauches as "test". nasa don't make "launch tests". apollo 1-7 are not as SN1-9, they were all known. apollo 1-9 will be more like comparable as SN25+--Dwalin (talk) 19:48, 10 December 2020 (UTC)

REVERTED EDITS

The last edits left the article in an illegible state - since it's in HIGH demand right now - I did a quick revert - on grounds that having something that people can actually read is a good thing! SteveBaker (talk) 19:52, 10 December 2020 (UTC)

Outcome
All test objectives achieved
Most test objectives achieved
Failure