Hacker News new | past | comments | ask | show | jobs | submit login
SpaceX rocket explodes during testing over Texas (bbc.co.uk)
143 points by richardwigley on Aug 23, 2014 | hide | past | web | favorite | 87 comments

BBC Updated the video to the 'better video' in @spacefight's link (the original was taken from a more distant vantage point).

SpaceX Statement in full (copied from facebook page):

Earlier today, in McGregor, Texas, SpaceX conducted a test flight of a three engine version of the F9R test vehicle (successor to Grasshopper). During the flight, an anomaly was detected in the vehicle and the flight termination system automatically terminated the mission.

Throughout the test and subsequent flight termination, the vehicle remained in the designated flight area. There were no injuries or near injuries. An FAA representative was present at all times.

With research and development projects, detecting vehicle anomalies during the testing is the purpose of the program. Today’s test was particularly complex, pushing the limits of the vehicle further than any previous test. As is our practice, the company will be reviewing the flight record details to learn more about the performance of the vehicle prior to our next test.

SpaceX will provide another update when the flight data has been fully analyzed.

> An FAA representative was present at all times.

Well thank goodness. I was so worried that a government agent was not present. All is well. Hail the FAA.

Is there any actual HD video? I find it really surprising I can't find any...

SpaceX will be in possession of some... Probably just not ready to release it yet.

I don't think they will release anything. Have they ever released any videos of their early failures?

Yes they have, for example http://youtu.be/0a_00nJ_Y88

Edit: and http://youtu.be/Qz0yJ8N3cA0

>>The unmanned rocket was destroyed when its self-destruct system was triggered after an unexplained malfunction.

It's interesting these rockets are equipped with self-destruct systems. What scenario would warrant a rocket purposely self-destruct?

Any number of reasons. Test ranges are usually in remote areas to avoid any immediate danger, but if the guidance or control systems malfunction, it can go somewhere unfortunate rather quickly.

I can't seem to find any good numbers, but it seems like the F9 rocket should have somewhere around 200 tons of fuel & oxidiser (kerosene & liquid oxygen). That, in addition to the few other hundred tons of rocket landing anywhere public is probably not going to end well.

IIRC, most rockets over a certain size are required to have some form of 'range safety package'[1] which may include a self-destruct capability for exactly these reasons.

In the course of the Challenger shuttle accident, the Solid Rocket Boosters were remotely destroyed by the range safety officer[2]

[1] https://en.wikipedia.org/wiki/Range_safety

[2] http://history.nasa.gov/rogersrep/v1ch9.htm

https://en.wikipedia.org/wiki/Intelsat_708 comes to mind as an example of when you'd want a failed launch to self-destruct. It went off course and landed in a village.

> It's interesting these rockets are equipped with self-destruct systems. What scenario would warrant a rocket purposely self-destruct?

When they're out of control and could hit someone.

From: http://en.wikipedia.org/wiki/Space_Shuttle_Challenger_disast...

> The more robustly constructed crew cabin also survived the breakup of the launch vehicle; while the SRBs were subsequently destroyed remotely by the Range Safety Officer, the detached cabin continued along a ballistic trajectory and was observed exiting the cloud of gases at T+75.237.

In the images, you can see the solid rocket booster (SRB) following a random path after the explosion.

Avoidance of cases like this: https://en.wikipedia.org/wiki/Intelsat_708

A video link to the actual launch failure https://www.youtube.com/watch?v=yfMbGPf4r9g

Not only do these rockets have them, but manned craft like the Space Shuttle are destructible from the ground (one hopes the encryption keys are well-protected). You don't want an out-of-control bomb heading towards Orlando.


Actually, your link states that the orbiter wasn't destructible.

It wasn't much of a bomb in comparison to the rest of the components.

Honestly, play KSP. Once you flip a rocket, its nearly unrecoverable and it just accelerates into the ground.

Not only that, but it accelerates at a higher rate because the thrust vector and gravitational pull are pointing in the same direction.

I wonder if the reusable rockets from SpaceX could recover from a flipped position during re-entry?

No, they can't. Once you're supersonic, the aerodynamic stress will tear a rocket into shreds if there's even a few degrees of angle of attack.

But if you're high enough, the forces are not very high, so there's probably a Q limit, (that is, proportional to density * speed^2).

Yeah, the landing attempts obviously flip before entry, while they're out of the atmosphere.

Because if they lose guidance but the motor keeps running the thing could hit a populated area.

Interestingly the Russian space program never used range safety systems, but then they had an empty part of Kazakhstan to launch in.

On the other hand, they had a system on their unmanned spacecraft that blew them up with TNT if there was a chance of them coming down outside the USSR. It was regularly triggered during their heyday in the 60's. There was a lot of gnashing of teeth from the engineers when that prevented them from collecting data on failed missions.

I read about it in Boris Chertok's memoirs (http://www.nasa.gov/home/hqnews/2012/feb/HQ_12-054_Chertok_R...), which are huge and sprawling, he was there from the beginning, from the mission to get V2 stuff from Germany after WW2, through the Soviet Union's fall.

Great video of an example:


Then imagine if it was headed for somebody's house (or a city). You'd want to blow it up in the air before it reached the ground.

I absolutely love that video - it's pretty much Kerbal Space Program in real life!

I agree. I'm glad nobody got hurt, so I can enjoy it without guilt.

My favorite part is at the end when the sound from the explosion arrives. Everything is calm, then suddenly KABOOM! Really drives home how far away the thing is (because it takes so long for the sound to arrive) and how big it is (because it's still a massive KABOOM even at that distance).

the KABOOM is impressive, shock wave at that distance is more telling

A rocket going far off the designated trajectory. It would be a catastrophe for a space launch rocket to fall into a civilian area.

probably isn't such a difficult system to add when you're adding that capability to a massive tank of pressurized fuel.

One can imagine all sorts of instances where it'd be preferable to litter the area with debris rather than missle-ing the rocket into a densely populated area.

Alien on board comes to mind.

"We've got to tunnel one of these vehicles into the ground by trying something really hard." - Gwynne Shotwell, President and COO of SpaceX

This puts "move fast and break things" into a whole new perspective.

I'm glad I can joke about this because nobody was hurt. Saw the headline at the top of the frontage and it immediately brought back memories of seeing Challenger on the news when I was 9, which rather changed my perspective on space travel at the time.

Indeed, but when I move fast and break things, my website goes down, and maybe costs a few thousand in lost sales. In terms of man hours and construction, this broken thing must cost tens of millions! That's some agile.


How NOT to do it. Not that China even bothered to check how many died, apparently.

please explain

As an experimental vehicle, you're generally supposed to be pushing the limits of what it can do to find those limits. Essentially, they're saying "no accidents means we aren't trying hard enough to break the thing".

Yes. If you read interviews, they had stuff like Elon having complaints that they aren't pushing the envelope hard enough since they don't get any crashes.

Maybe they tested the control system's abilities by pushing it over very quickly.

I bet there's people at SpaceX who were involved on the DC-X that did a successful translation to horizontal and back in 1995. https://www.youtube.com/watch?v=wv9n9Casp1o

Hope SpaceX has enough spare hardware that they can get quickly back up to testing again. I don't see any reason they wouldn't since they have produced quite many leg sets already - maybe they were planning to moving to newer hardware iterations anyway and the vehicle was "expendable", or at least the risk with dangerous tests was more acceptable.

It's interesting how diametrically different are the mindsets of engineers in one hand and the general public on the other. When I saw the video I thought it was a very bad news for SpaceX's name and I imagine some people might think the same. Like that one time when a Tesla car got fire. I remember reading news that highlighted the dangers of electric cars, etc., but then some commentators quickly pointed out the obvious: gasoline vehicles catch fire all the time.

But the point is, it's in benefit of everybody to get educated about engineering issues. I hope media outlets don't start going like "SpaceX: Too Dangerous? A rocket may land in your backyard". For me at least, it's interesting to learn that they WANT to break the thing to find the weak spots.

I see a difference between SpaceX and Tesla in this regard: Tesla had a production car catch on fire with a customer using it, while SpaceX broke their rocket doing an experiment with rockets during development.

It seems pretty reasonable you break your things while you're testing them out - I mean, we've all see those car crash commercials, right?

I also almost got a job there, and toured that facility during the final interview. Their testing at that site involves firing rockets that are secured to the ground, and monitoring the results from an underground bunker. I lived in that area up until this year, and I've never seen them launch anything from that site. I didn't even know they had that infrastructure there! There are also houses all around that facility, so I am surprised to see them launch this thing.

I almost expect a statement that this was a successful test of the self-destruct mechanism. ... Oops, looks like they really did it.

Huh. Why did the explosion produce two distinct, different-colored smoke clouds?

Also, what's the environmental impact of this kind of accident?

The rocket has two tanks, one with kerosene, one with liquid oxygen. I bet the black fireball is the kerosene burning, the white cloud is the liquid oxygen vaporizing.

Environmental impact is likely small, the kerosene would have burned anyway and there are little other hazardous chemicals on board. (Unlike for example the Proton failure linked by kryptisk above which is fueled with really nasty hypergolic fuels.)

To add to that. I'm sure Space X attempts to collect all the debris that falls to the ground.

I hope there's a way to disable the self-destruct mechanism when they start doing manned flights!

There won't be. If the self-destruct triggers, then the rocket is 1) already going to cause their deaths, and 2) going to potentially save the lives of people on the ground.

Now, that said, this is why we have escape towers on pods (and the escape rockets around the sides of the Dragon 2). In the case of an emergency, those eject the pod from the rocket so that it can be returned safely, even as the rocket is destroyed.

The shuttle had its own variant although that was manual, not automatic http://www.popularmechanics.com/science/space/nasa/4262479

I believe it doesn't matter what the craft is, there is always some manner of self-destruct to ensure a larger population isn't threatened by debris from an exploding rocket.

It does matter. I doubt you'll load 300 willing passengers on a 747 if it was widely known it had a self destruct.

I would guess each case has to weigh its pros and cons for inclusion of a system. There is such thing as acceptable risk to population. If FAA mandated them carte blanche, Michael Bay movies would become documentaries.

A 747 isn't a spacecraft travelling at supersonic speeds, so it's not really the type of craft relevant to the discussion. Cars, buses and boats also aren't loaded with self-destruct mechanisms for obvious reasons.

Obvious enough that I had assumed craft = aircraft as well since you can "take the keys out" of ground/water vehicles. I was challenging the thought that every aircraft needs self destruct, as I suspect it's a minority. If as you say supersonic capability is the threat threshold, fighter jets/rockets would have them but a C-5 Galaxy and Blackhawk might not (I don't know, I'm speculating).

In this context, I meant craft = spacecraft. Our current space launch technology relies on extremely volatile and dangerous fuel. Until that changes, a self destruct will be found in all large launch vehicles[1]


The threshold is more "Is it an unstable flying fuel tank* ? y/n"

*or solid block of fuel

I guess, military personnel (that astronauts so far have basically been) are less skittish in that regard than civilians.

Especially since it is understood that space travel is dangerous, while air travel is regarded as safe.

According to Neil Armstrong, Apollo 11 only had a "50–50 chance" of landing safely on the Moon's surface. They're willing to take enormous risks.

That might be out of context. There were a whole lot of points where the mission could have been safely aborted had any parameter been unacceptable. Look at Apollo 10, which was identical to 11 except for turning around when the lander was a few km above the surface. Failure to land does not mean a risk to safely. The risk wasn't survival, it was failing to complete the primary mission objective.

You might be right. Apparently, he thought they had a 90% chance of getting back safely to Earth.

It's one of the things you have to take into account for being one of only twelve human beings to walk on the surface of the moon.

I'd love to think that I would be able to do it, and whenever I get the slightest skittish on a plane, I laugh that I don't think I would hack it.

Dragon V2 will have a Launch Escape System, which is used to quickly separate the crew module from the rest of the rocket in case of an emergency.


The capsule sits on top of the rocket, and has the ability to quickly fly away if something goes wrong, whether it's the rocket exploding because it broke or because it's commanded to. There's nothing really incompatible with a manned rocket having a self-destruct system.

Very unlikely. The Space Shuttle also had this.


"the flight termination system automatically terminated the mission"... by blowing itself up?

How else would you terminate an active rocket flying through the air?

Land it safely on the ground 20 or 30 feet from where it took off, obviously


No, that would be "completing" the mission, not "terminating" it... ;-)

Yes, "flight termination system" refers to a bunch of explosives and a widget to blow them up on command.

Shit happens. Especially when you have a requirement for US citizenship for your employees...

Because nearly exclusively using US citizens has turned out so terribly for NASA the last 50 years.

That is factually incorrect. You do not have to be a US citizen to work there. You do, however, with few exceptions, have to be a US permanent resident.

And those rules are the US government's, not SpaceX's.

The Falcon 9 explodes again? ...While the European Vega rocket (aimed at the same size of payloads, but much more versatile of the Falcon 9[1]) is 3 success out of 3: http://en.wikipedia.org/wiki/Vega_rocket

The political will to create a private market for rockets is strong in US. So strong that a lot of American public money has been used to save SpaxeX from failure, multiple time.

But concretely, what's the point of it? NASA worked fairly well if well founded. Now for political reasons they prefer to give money to SpaceX, even if they have issues with their job...

[1] Vega launch satellites to Sun synchronous orbit, satellites here are earth observers and usually a few tonnes at most, the majority much smaller. Launches here are very infrequent, so matching spacecraft is difficult. Falcon 9 isn't currently equipped with an appropriate fairing to do dual launch like this without satellites built to stack (which most aren't). Falcon 9 might haul 9 tonnes, but you might not be able to find 9 tonnes of spacecraft to that orbit. Vega payloads are also kept in a clean environment and fairing, something SpaceX hasn't demonstrated. Horizontal integration can be an issue. ITAR issues... The list goes on.

Are you trolling? Most rocket platforms have had a rate of failure. The ESA's Ariane 5 (predecessor of the VEGA rocket) had launch failures as well. SpaceX has had an amazing run in comparison to most other platforms.

>But concretely, what's the point of it? NASA worked fairly well if well founded...

NASA doesn't build rockets. They have always outsourced the job to aerospace companies.

> Now for political reasons they prefer to give money to SpaceX

I think you'll find that political reasons are keeping NASA from buying more launches from SpaceX.


EDIT - changed my wording from "very checkered history" to "had launch failures as well" - in my rush to respond, I went from memory and checking the facts found my words too strong.

> The ESA's Ariane 5 (predecessor of the VEGA rocket) had a very checkered history.

Ariane 5 is not the predecessor of Vega. They are two very different project, for different payloads. And look at the table of success and failure rate of the Ariane 5 itself http://en.wikipedia.org/wiki/Arianne_5

> SpaceX has had an amazing run in comparison to most other platforms.

Clearly that's not the case.

> I think you'll find that political reasons are keeping NASA from buying more launches from SpaceX.

From that interview: http://www.cbsnews.com/news/tesla-and-spacex-elon-musks-indu...

Elon Musk: NASA called and told us that we'd won a $1.5 billion contract. And I couldn't even hold the phone, I just blurted out, "I love you guys."

Scott Pelley: They saved you.

Elon Musk: Yeah, they did.

> Ariane 5 is not the predecessor of Vega.

I stand corrected. The point is that the ESA's rocket program is far from flawless as your "3 from 3" post above suggests.

>> SpaceX has had an amazing run in comparison to most other platforms.

> Clearly that's not the case.

From the article, this was a TEST flight. Exploding rockets is an expected part of the development process.

In that context, NASA was the customer. They didn't bail them out. It wasn't charity.

The Falcon 9 never exploded. So far it has made 11 successful flights to orbit, it lost an engine mid-flight, once, Grasshopper made 8 successful hops, and G9R Dev1, a three engined first stage with grid fins, self-destructed on its 5th flight, probably while trying to do something no rocket has ever done. Note that SpaceX president Gwynne Shotwell said last year that she feels the engineers are not pushing the reusability testing enough because there were no failures yet.

NASA is a terrible rocket design bureau good only for pork barrel, which is what is has been doing for decades. NASA had to admit that developing Falcon 9 (100% privately funded by SpaceX and private investors) would have cost more than twice if it was done by NASA (http://www.nasa.gov/pdf/586023main_8-3-11_NAFCOM.pdf).

The point of it is that the US government finds it politically inexpedient and costly having to pay Russia or United Launch Alliance[1] (who sometimes use Russian engines anyway[2]) in order to get people and equipment to space. SpaceX is an obvious path away from the Russia Space Agency / ULA.

For things that the government wants to launch from the US, having a second US-based option for space launch will get the US out of the bilateral monopoly that exists with ULA.

[1] http://en.wikipedia.org/wiki/United_Launch_Alliance [2] http://spaceflightnow.com/news/n1408/20rd180delivery/

You uh, do realize that this was a test-bed rocket right? Designed to do experimental things?

Also, I was not aware of any Falcon 9 exploding previously. What on earth are you talking about?

Even I agree this might just be a experimental? for which they might not have permission for. or may be i am speculating

This is so incorrect it's almost funny. For the record:

First: Falcon 9 so far has 11/11 successful launches. This was not a Falcon 9, this was a test flight of a sub-orbital development vehicle (F9R-"Dev").

Second: "Vega rocket (aimed at the same size of payloads...". Your Wikipedia link says maximum payload to LEO is 1,963 kg, while Wikipedia quotes Falcon9's maximum payload to LEO is 13,150 kg.

And how much do they charge for those 1963kg? Wikipedia quotes 32 million euros per launch, or $42M, corresponding to $21,000/kg, for the Vega, and $4,109/kg for the F9...

This was an auto-destruct, it didn't just randomly explode due to engine failure. Some other parameters of the test launch were either incorrectly configured or went out of bounds and the flight computer destroyed it. Rockets are hard, sometimes you have to fail fast.

Well actually NASA, like any beaurocracy, grew so large that rockets were costing billions of dollars to launch a new rocket. Without market forces to keep the costs in check, no one thought this a problem and contracts were issued based on paperwork driven development.

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact