SpaceX Starship - Orbital Test Flight - Starbase TX

Launch Date: April 20
Launch Window: 8:28am CDT (6:28am PDT, 13:28 UTC) - 62 minute window
Launch site: LC-1? - Starbase, Boca Chica Beach, Texas
Core Booster Recovery: Expended in Gulf
Starship Recovery: Water landing near Hawaii
Booster: Super Heavy Booster 7
Starship: Starship 24
Mass: No mass simulator mentioned
Orbit: LEO-ish
Yearly Launch Number: 26

A SpaceX Super Heavy and Starship launch vehicle will launch on its first orbital test flight. The mission will attempt to travel around the world for nearly one full orbit, resulting in a re-entry and splashdown of the Starship near Hawaii.

Webcast:

 

[mongo]

That seems for sure. I don't understand how SpaceX engineers modeling the forces and heat those 33 engines would generate and for how long, would not have concluded the pad was way underprotected. The substantial launch successes might just as well have gone against SpaceX.

Because they extrapolated from the erosion caused by the 31 engine 50% thrust static fire, not knowing the higher thrust of launch would result in a different failure mode?

https://twitter.com/x/status/1649800747834392580

 

[adiggs]

There's just one thing about the whole flame trench thing that has me bamfoozled. I'm totally on board with how Elon ordinarily works, with iterate fast, test often, and so forth.

That has me bamfoozled is -- what's the up side? If it turns out that they don't need the flame trench, does it cut the cost of building the launch pad / facility in half (or alternatively, cost 2x as much with the flame trench than without)? Clearly there is some level of incremental work, and therefore cost, but it doesn't seem to me like a flame trench would have all THAT great of an impact on cost. Does the tower need deeper footings when its straddling the trench?


I like doing risky tests, with frequency. You learn faster, and the stuff you learn is more solidly known and understood. But the risky tests need to be pushing something meaningful forward. The absence of flame trench when it seems like the need for one is so well understood suggests to me that they are pretty expensive to build and/or refurbish and/or maintain; that's the only thing that makes sense to me (outside interested observer; I was a software engineer though, so its most likely I have no clue

If the trench is as close to a 1-time cost, and as cheap to build, as I think they would be, then trying to do without makes no sense. Suggesting they are a lot more expensive than I think.

 

[Knightshade]

IIRC the issue is they'd need to do a bunch of environmental impact type stuff for the trench

(there's also the issue of Elons expecting to want to be launching a LOT of these things, often, in the future, so the simpler/cheaper you can make the MANY launch pads they'll need the better.... but personally I'm kinda with you in that I'd rather OVERbuild the FIRST launch pad-- to get the actual vehicle working correctly-- and then if data tells me I can build the NEXT launch pad (or next 10 or 100 of them) with LESS stuff-- awesome....but I'm no rocket surgeon)

 

[Dan D.]

IIRC the issue is they'd need to do a bunch of environmental impact type stuff for the trench

(there's also the issue of Elons expecting to want to be launching a LOT of these things, often, in the future, so the simpler/cheaper you can make the MANY launch pads they'll need the better.... but personally I'm kinda with you in that I'd rather OVERbuild the FIRST launch pad-- to get the actual vehicle working correctly-- and then if data tells me I can build the NEXT launch pad (or next 10 or 100 of them) with LESS stuff-- awesome....but I'm no rocket surgeon)

What about using no concrete at all for the pad material, just a big ol' pre-cratered dirt pad with a much better reinforced concrete/steel stand - perhaps twice as high up from the ground. Sure the rocket will churn up and blast away dirt. Pre-sifted hard-packed dirt, so there are no chunks of anything substantial. Surround the outer perimeter with metal mesh fences of decreasing particle size to filter out the dirt cloud, and a water deluge for the sound dampening, and a water spray to reduce the dirt cloud.

Isn't the best launch pad material, no launch pad material?

Can it not be launched from much higher, say 100m above the ground on a super-tall stand?

 

[mongo]

There's just one thing about the whole flame trench thing that has me bamfoozled. I'm totally on board with how Elon ordinarily works, with iterate fast, test often, and so forth.

That has me bamfoozled is -- what's the up side? If it turns out that they don't need the flame trench, does it cut the cost of building the launch pad / facility in half (or alternatively, cost 2x as much with the flame trench than without)? Clearly there is some level of incremental work, and therefore cost, but it doesn't seem to me like a flame trench would have all THAT great of an impact on cost. Does the tower need deeper footings when its straddling the trench?


I like doing risky tests, with frequency. You learn faster, and the stuff you learn is more solidly known and understood. But the risky tests need to be pushing something meaningful forward. The absence of flame trench when it seems like the need for one is so well understood suggests to me that they are pretty expensive to build and/or refurbish and/or maintain; that's the only thing that makes sense to me (outside interested observer; I was a software engineer though, so its most likely I have no clue

If the trench is as close to a 1-time cost, and as cheap to build, as I think they would be, then trying to do without makes no sense. Suggesting they are a lot more expensive than I think.

Part of the issue is Boca Chica starts out 3 feet above sea level. To make a trench, you either need to dig below the water table and somehow keep it pumped out with a ramping exit, go NASA on it and haul in enough fill to create a mound with a void in the center, or raise the launch mount even higher relative to the ground and build a flame wall braced for takeoff thrust.
First is impractical, second takes a lot of time, dirt, and area (slope of mound), third requires an even bigger crane to put everything together. Although, the OLM design is sort of an open, undirected, trench (sans deluge).

IIRC the issue is they'd need to do a bunch of environmental impact type stuff for the trench

(there's also the issue of Elons expecting to want to be launching a LOT of these things, often, in the future, so the simpler/cheaper you can make the MANY launch pads they'll need the better.... but personally I'm kinda with you in that I'd rather OVERbuild the FIRST launch pad-- to get the actual vehicle working correctly-- and then if data tells me I can build the NEXT launch pad (or next 10 or 100 of them) with LESS stuff-- awesome....but I'm no rocket surgeon)

The problem I see is, if you develop on a theoretically over-designed system, who is going to take the risk to build a lesser stage zero for production flight articles? Things, once designed in, are hard to design out.

 

[scaesare]

Because they extrapolated from the erosion caused by the 31 engine 50% thrust static fire, not knowing the higher thrust of launch would result in a different failure mode?

https://twitter.com/x/status/1649800747834392580

Indeed.

The hard part of doing something that's never been done before is that.... it's never been done before.

 

[Dan D.]

Because they extrapolated from the erosion caused by the 31 engine 50% thrust static fire, not knowing the higher thrust of launch would result in a different failure mode?

https://twitter.com/x/status/1649800747834392580

Seems like they could have tested the launch pad using 33 raptors in a sustained static fire test at full thrust in isolation.

Also a good way to test trenches and water deluge systems.

Please don't let me hear that the next time they test the launch pad (redesign) is going to be at the time of another orbital launch. Can they not refine the design first?

 

[JB47394]

Seems like they could have tested the launch pad using 33 raptors in a sustained static fire test at full thrust in isolation.

Unless they figured that the pad would be so damaged by erosive forces (their accepted mode of failure) that they'd have to replace the Fondag for the launch, which is both expensive and time consuming. Their data said that it should survive, so they took a chance. In hindsight, that was a big mistake, but SpaceX takes lots of risks and embraces novel solutions. The fact that they've been burned as few times as they have is a testament to the abilities of their people.

 

[mongo]

Seems like they could have tested the launch pad using 33 raptors in a sustained static fire test at full thrust in isolation.

Also a good way to test trenches and water deluge systems.

Please don't let me hear that the next time they test the launch pad (redesign) is going to be at the time of another orbital launch. Can they not refine the design first?

Wouldn't that test have still destroyed stage zero and potentially the stack without the benefit of flight data?
From the sounds of it, the OLM clamps don't hold the booster down for launch. Once thrust to weight (TWR) is over 1, it's liftoff. If so, they can't do a full thrust static fire (66% max thrust assuming 1.5 TWR), and a booster only test is even more limited (without a >2 million pound mass simulator) which is likely part of the reason the 31 engine test was at 50%.

 

[Dan D.]

Wouldn't that test have still destroyed stage zero and potentially the stack without the benefit of flight data?
From the sounds of it, the OLM clamps don't hold the booster down for launch. Once thrust to weight (TWR) is over 1, it's liftoff. If so, they can't do a full thrust static fire (66% max thrust assuming 1.5 TWR), and a booster only test is even more limited (without a >2 million pound mass simulator) which is likely part of the reason the 31 engine test was at 50%.

Yes of course doing a static fire test of launchpad designs could have destroyed the testbed stage zero, that's the point of the test. But it wouldn't have destroyed the actual stage zero, the orbital Starship, nor anything not related to the 33 raptors and their fuel systems. The testbed stage zero would be located elsewhere to the actual stage zero, perhaps in the middle of nowhere.

Especially now, when we know the launchpad design doesn't work, and their forecasting didn't work. They need to build something that does work. Test it a few times to be sure it works, don't say "here's another experimental idea, we'll test it for the first time at the next valuable orbital launch".

Full thrust static fire isn't a problem if your testbed is STATIC, ie: a reinforced stand and fixed booster segment that can adequately withstand the forces; you don't do it with a real booster stage.

TL;DR: SpaceX needs to TEST experimental ideas until they work fully. No launch diverter and no water deluge was an experimental idea that they didn't test, and it failed. The lesson learned should be to TEST things like that, before they are used in a live launch.

 

[mongo]

Yes of course doing a static fire test of launchpad designs could have destroyed the testbed stage zero, that's the point of the test. But it wouldn't have destroyed the actual stage zero, the orbital Starship, nor anything not related to the 33 raptors and their fuel systems. The testbed stage zero would be located elsewhere to the actual stage zero, perhaps in the middle of nowhere.

Especially now, when we know the launchpad design doesn't work, and their forecasting didn't work. They need to build something that does work. Test it a few times to be sure it works, don't say "here's another experimental idea, we'll test it for the first time at the next valuable orbital launch".

Full thrust static fire isn't a problem if your testbed is STATIC, ie: a reinforced stand and fixed booster segment that can adequately withstand the forces; you don't do it with a real booster stage.

TL;DR: SpaceX needs to TEST experimental ideas until they work fully. No launch diverter and no water deluge was an experimental idea that they didn't test, and it failed. The lesson learned should be to TEST things like that, before they are used in a live launch.

So build twice as many OLMs/OLTs with twice as many sets of GSE in twice as many locations, so that they potentially avoid needing to repair one specific set of the above?
Boca Chica is the test site. Canaveral is the long term, external party payload, site.

As planned, the next attempt will be with the steel water cooled plate installed.

 

[Electroman]

Yes of course doing a static fire test of launchpad designs could have destroyed the testbed stage zero, that's the point of the test. But it wouldn't have destroyed the actual stage zero, the orbital Starship, nor anything not related to the 33 raptors and their fuel systems. The testbed stage zero would be located elsewhere to the actual stage zero, perhaps in the middle of nowhere.

Especially now, when we know the launchpad design doesn't work, and their forecasting didn't work. They need to build something that does work. Test it a few times to be sure it works, don't say "here's another experimental idea, we'll test it for the first time at the next valuable orbital launch".

Full thrust static fire isn't a problem if your testbed is STATIC, ie: a reinforced stand and fixed booster segment that can adequately withstand the forces; you don't do it with a real booster stage.

TL;DR: SpaceX needs to TEST experimental ideas until they work fully. No launch diverter and no water deluge was an experimental idea that they didn't test, and it failed. The lesson learned should be to TEST things like that, before they are used in a live launch.

+1

 

[scaesare]

Yes of course doing a static fire test of launchpad designs could have destroyed the testbed stage zero, that's the point of the test. But it wouldn't have destroyed the actual stage zero, the orbital Starship, nor anything not related to the 33 raptors and their fuel systems. The testbed stage zero would be located elsewhere to the actual stage zero, perhaps in the middle of nowhere.

That version of the Booster was already deprecated, It uses hydraulics for gimbal actuation rather than the new electrics. No doubt there are 100's of other improvements since that version as well. SpaceX already has Rocket Gardens that are getting awfully crowded with articles that aren't capable of (or worth) flying. So SN24 was: flight worthy, disposable, and able to provide valuable data. Why not fly it?

Especially now, when we know the launchpad design doesn't work, and their forecasting didn't work. They need to build something that does work. Test it a few times to be sure it works, don't say "here's another experimental idea, we'll test it for the first time at the next valuable orbital launch".

What better way to test than under actual launch conditions... which is exactly what they did.

Full thrust static fire isn't a problem if your testbed is STATIC, ie: a reinforced stand and fixed booster segment that can adequately withstand the forces; you don't do it with a real booster stage.

Or... you do, if you have extra unneeded boosters around. Ones that can provide you additional flight data that a fixed test booster segment cannot.

TL;DR: SpaceX needs to TEST experimental ideas until they work fully. No launch diverter and no water deluge was an experimental idea that they didn't test, and it failed. The lesson learned should be to TEST things like that, before they are used in a live launch.

I suspect your use of "needs" here is a bit of an overextension as you are using it in an argument that this wasn't a valid test. Of course it was. This was exactly an "experimental idea". The now have results that inform them which theories didn't work (not needing flame management), and which did (their plumbing system successfully fed Raptors above 50%, their staged ignition system worked, Starship tile mounting survived >50% thrust launch, booster rigidity is capable of dealing with Starship loads, the OLM can withstand greater launch thrust, the list goes on and on....)

SpaceX didn't get to where it was taking the incremental steps rocket development historically has. "Fail Fast" is alive and well... and seems to be serving SpaceX well also...

 

[scaesare]

Seems like they could have tested the launch pad using 33 raptors in a sustained static fire test at full thrust in isolation.

Also a good way to test trenches and water deluge systems.

Please don't let me hear that the next time they test the launch pad (redesign) is going to be at the time of another orbital launch. Can they not refine the design first?

Why not?

If they have data from this test, giving them a greater degree of confidence it will work, and they have a booster that needs to be expended/retired anyway, why not test multiple things at once?

 

[Grendal]

EA's newest video shows a lot of detail.

 

[Dan D.]

SpaceX didn't get to where it was taking the incremental steps rocket development historically has. "Fail Fast" is alive and well... and seems to be serving SpaceX well also...

You ask a lot of good questions in response to my questions of the launch, which just shows we have a different response to the test method SpaceX is using here. They support fast and fail often, and I support methodical and less reckless. Different strokes for different folks, so I'm not going to respond to your points - as they are valid to a different viewpoint than mine.

The one thing that SpaceX risks is time. They really need to get this program moving forward more quickly, this latest launch mishap now seems likely to cost them months or years in delay. If only they didn't have so many 'successful' disasters...


One viewpoint on this launch attempt:
"The FAA is going to have a difficult decision on its hands. Do they let SpaceX test out an unproven heat and sound plume protection device, given the last test of a 'new way to launch super rockets' failed miserably in a hailstorm of concrete? If the FAA approves this protection equipment and no other regulatory process slows down SpaceX, we could see a launch in 6-12 months.

However, if the FAA insists on a full civil rebuild to accommodate a traditional flame trench, the Corps permitting alone, coupled with a Texas Commission on Environmental Quality (TCEQ) Clean Water Act 401 review, will take at least 2-3 years before construction can begin."

 

[scaesare]

You ask a lot of good questions in response to my questions of the launch, which just shows we have a different response to the test method SpaceX is using here. They support fast and fail often, and I support methodical and less reckless. Different strokes for different folks, so I'm not going to respond to your points - as they are valid to a different viewpoint than mine.

The one thing that SpaceX risks is time. They really need to get this program moving forward more quickly, this latest launch mishap now seems likely to cost them months or years in delay. If only they didn't have so many 'successful' disasters...


One viewpoint on this launch attempt:
"The FAA is going to have a difficult decision on its hands. Do they let SpaceX test out an unproven heat and sound plume protection device, given the last test of a 'new way to launch super rockets' failed miserably in a hailstorm of concrete? If the FAA approves this protection equipment and no other regulatory process slows down SpaceX, we could see a launch in 6-12 months.

However, if the FAA insists on a full civil rebuild to accommodate a traditional flame trench, the Corps permitting alone, coupled with a Texas Commission on Environmental Quality (TCEQ) Clean Water Act 401 review, will take at least 2-3 years before construction can begin."

Agree on the agree to disagree based on perspective .

But I have to ask:

the one thing that SpaceX risks is time. They really need to get this program moving forward more quickly

Really??

SpaceX has gone from startup barely managing to get Falcon 1 off the launchpad on it's final try in 2008 to developing reusable rockets to building the first production full-flow-staged combustion rocket engine on the planet to testing the most powerful rocket ever built in not much longer than far more established players take to build a single vehicle using many pre-existing systems and parts (SLS anyone?)

Give me an example of anyone moving more quickly?

 

[SO16]

Agree on the agree to disagree based on perspective .

But I have to ask:



Really??

SpaceX has gone from startup barely managing to get Falcon 1 off the launchpad on it's final try in 2008 to developing reusable rockets to building the first production full-flow-staged combustion rocket engine on the planet to testing the most powerful rocket ever built in not much longer than far more established players take to build a single vehicle using many pre-existing systems and parts (SLS anyone?)

Give me an example of anyone moving more quickly?

Also NASA is moving at the speed they are with much more backing from the US govt than SpaceX.

 

[Unclepaul]

I believe the reason Space X went forward with this test was time. They ran a 50% static test first and thought that the full power event would significantly wear the launch pad, but not destroy it. OOPS!

Rocket was the most powerful launch the World had ever seen. They chose to go ahead and see what would happen. Now they know.

Imagine that with this knowledge they will be able to construct a state of the art, reusable launch site. Will include more concrete, stronger support, a flame trench, water quench system, plus heavy steel plates that can be replaced quickly when worn. They may also need to move to a different site, or even to choose to launch from an over water platform. Earlier drawings showed StarShip launching from a water base.

 

[mongo]

You ask a lot of good questions in response to my questions of the launch, which just shows we have a different response to the test method SpaceX is using here. They support fast and fail often, and I support methodical and less reckless. Different strokes for different folks, so I'm not going to respond to your points - as they are valid to a different viewpoint than mine.

The one thing that SpaceX risks is time. They really need to get this program moving forward more quickly, this latest launch mishap now seems likely to cost them months or years in delay. If only they didn't have so many 'successful' disasters...


One viewpoint on this launch attempt:
"The FAA is going to have a difficult decision on its hands. Do they let SpaceX test out an unproven heat and sound plume protection device, given the last test of a 'new way to launch super rockets' failed miserably in a hailstorm of concrete? If the FAA approves this protection equipment and no other regulatory process slows down SpaceX, we could see a launch in 6-12 months.

However, if the FAA insists on a full civil rebuild to accommodate a traditional flame trench, the Corps permitting alone, coupled with a Texas Commission on Environmental Quality (TCEQ) Clean Water Act 401 review, will take at least 2-3 years before construction can begin."

How would timing be any faster, or FAA any more permissive if SpaceX built an entirely separate site?

Yes of course doing a static fire test of launchpad designs could have destroyed the testbed stage zero, that's the point of the test. But it wouldn't have destroyed the actual stage zero, the orbital Starship, nor anything not related to the 33 raptors and their fuel systems. The testbed stage zero would be located elsewhere to the actual stage zero, perhaps in the middle of nowhere.

Especially now, when we know the launchpad design doesn't work, and their forecasting didn't work. They need to build something that does work. Test it a few times to be sure it works, don't say "here's another experimental idea, we'll test it for the first time at the next valuable orbital launch".