totm nov 2023 SpaceX Discussion Thread

[Minmus Taster]

10 minutes ago, darthgently said:

To keep things science-based, it is plausible that SpaceX isn’t failing enough to optimize iterative learning according to experts in learning

 

An interesting proposition, but rocketry is expensive and even SpaceX can't afford to expend so many vehicles on failed flights. Sooner or later you need to see a return on investment, though for them that day seems to be relatively far off.

[darthgently]

3 minutes ago, Minmus Taster said:

An interesting proposition, but rocketry is expensive and even SpaceX can't afford to expend so many vehicles on failed flights. Sooner or later you need to see a return on investment, though for them that day seems to be relatively far off.

Exactly.  There is likely an optimization of a cost and learning function that minimizes cost while maximizing iterative learning and I wouldn’t be shocked if SpaceX was in that ballpark.

[Minmus Taster]

[Nuke]

5 hours ago, SunlitZelkova said:

I think the only reason the N1 didn’t work was because of lack of resources and political support, two things SpaceX now has (the latter in the form of internal backing + HLS contract). The N1 nearly made it to orbit on the 4th flight and probably would have succeeded on the 5th.

But yes, stuff like the Nedelin disaster had no benefit and was pretty easily avoidable.

I was planning to say this in response to the hubbub about lack of regulations and accusations of negligence but I didn’t: SpaceX has no incentive to build an unsafe rocket. They’re a business that doesn’t have a lot of leeway in terms of fatalities or injuries. A carmaker can build something that kills 1,000 people a year and people think it’s normal, but a rocket maker gets one guy killed for the first time in seven years and accusations the company is pure evil go flying.

This is extreme, but even without the FAA (or with a reduced FAA) I would trust SpaceX to at least do some due diligence to make sure people are out of harms way before launching. I also trust them to figure out what went wrong and try to fix it before the next flight.

They’re just not building a minivan where 1,000 out of a million can fail and people won’t care. Space is high profile stuff and they have enormous incentive to improve safety even without having regulators looking over their shoulders all the time.

I think that’s saying a lot considering I’m a person who on the other hand didn’t, and still doesn’t trust SpaceX to do due diligence to protect the environment around Starbase (although as much as I lack that trust I don’t bring it up anymore because I’ve just accepted Starbase is a critical national security facility now and nothing can realistically be done).

proportionality needs to be observed. a thousand people dying a year in auto accidents sounds bad, but then again how many cars are there? also with cars when there is an accident, you can usually blame one driver or the other, and the car maker gets a pass. usually also the people who designed the highway system and i am convinced some intersections are defacto death camps. if a million people fly in rockets then a couple deaths dont look so bad. every time a plane crashes the body count is usually near or in the hundreds, yet its the safest way to travel because of the shear volume of air traffic. idk how it lapped trains and steamers in terms of safety, perhaps that statistic is just airline marketing. probibly because there is a history of shipping going back to the age of sail that was not always as safe as it is now. seems train wrecks never happen either, ive heard about a dozen aircraft incidents to one train incident. trains being close to the ground are somewhat more survivable and you dont usually drown or catch fire when they derail. what is the lethality of those? 

maybe im wrong to bring up the lethality of the shuttle too often. seems everything we humans do comes with an element of risk, and its not really my place to cover everyone in bubble wrap, take away their sharp objects and tell them not to have fun to make myself feel better. society likes to drive cars, live in cities, work dangerous jobs, climb mountains, and go to mars (crossing the street belongs somewhere in that list). now that i think about it safetyism probibly does  more harm than good. we can play it safe until the sun explodes or the biosphere becomes uninhabitable either by our own hand or external forces out of our control. that's not how you build a galactic empire.

Edited Sunday at 04:02 AM by Nuke

[RCgothic]

Trains are safer in terms of fatalities per trip. 

Planes are safer in fatalities per mile. 

The difference is that planes traverse vastly larger distances on each trip than the typical train journey.

If you could make the same trip by either mode of transport, trains would be the safer option.

[magnemoe]

8 hours ago, Minmus Taster said:

An interesting proposition, but rocketry is expensive and even SpaceX can't afford to expend so many vehicles on failed flights. Sooner or later you need to see a return on investment, though for them that day seems to be relatively far off.

Who is true, but things was going very well at flight 5 and 6, they was able to catch the booster and do soft touchdown with ship after reentry. 
Had flight 7 went as well as 6  they probably tried to put it up into orbit on flight 8 or 9. 

Now something I find troubling is how ship started spinning wildly once one vacuum engine went out rater than reducing trust on the other engines and compensating with of the sea level engines. 

[Lisias]

2 hours ago, magnemoe said:

Now something I find troubling is how ship started spinning wildly once one vacuum engine went out rater than reducing trust on the other engines and compensating with of the sea level engines. 

Because nobody configure the Abort button on KSP, I mean, SpaceX!

[tomf]

4 hours ago, RCgothic said:

Trains are safer in terms of fatalities per trip. 

Planes are safer in fatalities per mile. 

The difference is that planes traverse vastly larger distances on each trip than the typical train journey.

If you could make the same trip by either mode of transport, trains would be the safer option.

I expect it would be interestingly more complicated than that. My guess is that the risk of train accident on a given journey is proportional to the length of the journey, mostly due to signalling problems, poorly maintained track etc. The air travel risk however is pretty concentrated in takeoff and landing.

I expect there is a theoretical journey length where the per trip safety of aircraft overtakes trains again.

[darthgently]

3 hours ago, magnemoe said:

is how ship started spinning wildly once one vacuum engine went out rater than reducing trust on the other engines and compensating with of the sea level engines. 

This is interesting.  There was a deeper side effect that trashed sensors or code.  Or maybe valves just stuck open.  The RCS seemed to be trying to correct the spin so the software seemed to be getting some attitude data but I wonder if data that the fact that several engines were kaput at that point was available.  It is like it still considered things being under balanced thrust or something.  Maybe a secondary  thrust balance calculation based on attitude changes is needed if not there

[Minmus Taster]

Looks to me like a RUD instead of an FTS trigger. This is probably the upper half of the ship if I had to guess.

[JoeSchmuckatelli]

39 minutes ago, Minmus Taster said:

Looks

So cool that people can and will do stuff like that! 

... 

I'm aware that the "move fast and break stuff / best part is no part / only put stuff back on when it no longer works" ideology incorporates events like the last two failures, and that from failure can come success.  So I'm hoping we hear from SX or some of the deeper diving watchers about what the likely problem is and how they will overcome the engineering challenges. 

That's the interesting stuff. 

The only question in my mind is whether the fast pace of construction has 'baked in' problems. 

From a public relations standpoint it would be better for SX to figure out this aft-fire-RUD thing before flying again.  However from a sunk cost / get info from what you have standpoint - it might actually be better for SX to fly a faulty-but-already-built ship to scrap in the ocean rather than send it to the yard for disassembly. 

[Nuke]

9 hours ago, RCgothic said:

Trains are safer in terms of fatalities per trip. 

Planes are safer in fatalities per mile. 

The difference is that planes traverse vastly larger distances on each trip than the typical train journey.

If you could make the same trip by either mode of transport, trains would be the safer option.

i knew they were gerrymandering the statistics on this.

[tater]

 

[darthgently]

Interesting mostly informed speculations:

 

[darthgently]

OMG.  I have seen the light

 

What is this camera case glazing made of?

[AckSed]

I saw the Raptor bells 'twanging' like the RS-25 on startup. Damn.

[magnemoe]

6 hours ago, darthgently said:

This is interesting.  There was a deeper side effect that trashed sensors or code.  Or maybe valves just stuck open.  The RCS seemed to be trying to correct the spin so the software seemed to be getting some attitude data but I wonder if data that the fact that several engines were kaput at that point was available.  It is like it still considered things being under balanced thrust or something.  Maybe a secondary  thrust balance calculation based on attitude changes is needed if not there

My guess is that the lines stayed open as the signal cables was cut, but then why did it not close? It might be that closing it because lack of signal during launch is also an failure mode?

It also did not look like flight termination system activated, was it turned off? I understand its turned off at some point, it could also be that it did not derivatet enough from light path?

 

[Minmus Taster]

1 hour ago, darthgently said:

OMG.  I have seen the light

 

What is this camera case glazing made of?

The vibration of those engine nozzles is what catches my eye the most, I never thought of them as being flexible.

[Deddly]

12 hours ago, RCgothic said:

Trains are safer in terms of fatalities per trip. 

Planes are safer in fatalities per mile. 

The difference is that planes traverse vastly larger distances on each trip than the typical train journey.

If you could make the same trip by either mode of transport, trains would be the safer option.

Sorry to continue this aside for a moment, but that doesn't add up in my admittedly tired brain. If planes are safer per mile, it should be safer to travel by plane always.

[darthgently]

36 minutes ago, Minmus Taster said:

The vibration of those engine nozzles is what catches my eye the most, I never thought of them as being flexible.

This got me interested so I asked grok and found out smooth vs ribbed nozzles flex more because they are less rigid.  Like 10mm of flex versus 1 or 2mm.  Also much of that flex e see in this vid (looks more than 10mm) probably goes away once the reflected acoustic energy near the launch mount and ground diminishes.

 Some of that apparent flex may be optical distortion from heat but it does look structural and not merely optical

Edited Sunday at 08:25 PM by darthgently

[tater]

4 minutes ago, Deddly said:

Sorry to continue this aside for a moment, but that doesn't add up in my admittedly tired brain. If planes are safer per mile, it should be safer to travel by plane always.

A confounder here would be that a train might carry the same number of people as a plane, but a "fatal" train trip might include it hitting a car—say with 1 death (could be the engineer at the front of the train if the car/pedestrian deaths don't count for this math). Planes (airliners) only rarely have incidents with small %s of occupants dead—tends more towards all or nothing. Per "trip" is doing some odd work here as well. You'll see people get on a Metro North train at 125th Street, getting off at Grand Central. That's a trip of 1 stop, and at speeds so slow I don't think fatality is even possible short of falling under a train car. So the number of person-trips could get very inflated.

[Minmus Taster]

2 hours ago, tater said:

 

Does this vehicle have a name yet? I recall that Freedom was named before launch.

[tater]

 

[Minmus Taster]

[mikegarrison]

43 minutes ago, Deddly said:

Sorry to continue this aside for a moment, but that doesn't add up in my admittedly tired brain. If planes are safer per mile, it should be safer to travel by plane always.

There is a reason why it's not adding up for you. Train travel and plane travel and boat travel and car travel are not 100% fungible. Trains and cars/buses/trucks mostly are, so comparing them directly is easiest.

Planes are mostly at risk during landing/takeoff, so the number of flights is perhaps more important then the total miles flown. Meanwhile trains/cars are pretty much most vulnerable while moving, and almost certainly have more fatalities mid-trip than at (intended) start or finish.

So you are comparing things that are very unlike each other. More so than apples and oranges.

======

All that being said, there is basically no doubt that travel by train, (commercial) plane, or bus is much safer than the same trip by private car.

Edited Sunday at 09:03 PM by mikegarrison