Hacker News new | ask | show | jobs
by szoszon 1154 days ago
If there are 5 more of these being built, how they will incorporate anything they learned during this failed attempt? Isn't that just a waste of money?
4 comments

Considering that the failures don't appear to point to fundamental design flaws of the structure of the vehicle (if anything it kind of overperformed structurally with how long it just tumbled despite being a 100m tall, mostly hollow tube), they'll do it the same way you would fix/upgrade a car, by swapping out the bad parts.
Some parts are interchangeable.

The countdown process needs optimization. There are things missing and things which shouldn’t be there. We’re talking about speed of loading propellant, density of it, etc.

Then there’s software. So much software.

And a million other things. Rockets are complex, launching them doubly so.

also, to test the software you don't need to launch the rocket, so I'm not sure why you even mention that..
Tell that to the mars climate orbiter: https://en.m.wikipedia.org/wiki/Mars_Climate_Orbiter
What's the test fixture for realistic vibration oscillations of the support struts or combustion instability as the rocket breaks the sound barrier?

At some point, your in silico model needs real world data to validate it.

Vibration testing is a thing, you know.. I actually do a lot of work with it so..
Software is full of constants you can estimate but need to measure to be sure.

A few

- How long fuel takes to load

- How much is lost due to venting, how fast we need to recycle that to keep the ship on the pad ready to launch

- Actual engine startup times when many starting at once

- Actual thrust as the vechicle leaves the pad

- Actual response to control authority (how fast the ship changes course) for vectoring, gas thrusters, flaps

- Actual vibration measurements, how much measured acceleration is noise

Exactly. Falcon 9 took years and a couple dozen launches to optimize the countdown and flight profile to the current best in class general purpose medium weight launcher with the flight hardware remaining relatively constant (if for nothing else than certification reasons.)
Yeah, that's why you engineer, and test before trying to launch something like that. There won't be any other attempt this year - I can bet my lunch money on this.
> There won't be any other attempt this year

Possible, but not a sure bet by any means and one I'd happily take the other side of.

From watching the launch, my absolutely pulled from my butt guess is that something went wrong with the stage separation system, and that the explosion was deliberately triggered by the SpaceX team / triggered by automated guardrails when the mission didn't go according to parameters.

If that's true, then they can just fix whatever is wrong with the stage separation. And if it's something else, they can probably fix that too. Doesn't seem like it was anything unchangeably / structurally wrong with the entire craft.

There was a lot of wrong with the launch from the start - failed engines, destroyed pad, parts of the rocket flying off during the start. Yes, it's amazing that it went for as far as it did, but there is no way they can fix all of it quickly.
Everything you just listed is very probably all from a single root cause (the pad not being able to handle the force) and the fix could be as simple as using a bigger water deluge system / having a pad that isn't just a flat block of concrete / etc.
If they don't have the launchpad designed to not cause all of these issues, why do you think the rest is designed correctly?
Because they knew the launchpad wasn't entirely suitable in advance. And it still got off the pad anyway.
I not sure if the engines failed or the inlet system had issues with pressure and couldn't feed the engines properly.
Yes, stage separation failed apparently. Then the flight termination system was triggered.
No only are there more being built, but this launch used engines that are already obsolete. They are learning stuff by building the rocket, by testing the components and by using them to do a test launch. They are trying to wring as much learning out of each step as possible.