Major Proton launch failure

[sal_vager]

Wow, this thread is pretty far from science, I better move it.

Oh and thank gods they used a probe core....

[rasheed]

Happens to me once in a while.......

The strange part is that the ASAS is not working well.

But Squad is fixing it

[OdinYggd]

Russia Today is reporting a "nuclear fuel" leak.

You sure it was nuclear? don't see any mention of it in that article.

Although the UDMH (Hydrazine) fuel that rocket was using is highly toxic, what didn't get consumed in the blast should disperse fairly reliably on its own.

Bigger question is did the fault originate as a control system malfunction, or was it an engine malfunction that resulted in a loss of control.

The video footage available is ambiguous, as the craft starts to rotate and veer off course mere moments before a visible plume of black smoke came out of one of the engine bays. It initially looked like it had saved itself and the smoke cleared, but then it went over on the other side and the smoke plume appeared again before arcing over and crashing.

That dark colored cloud could easily have been unburnt or partially burnt fuel being passed by an engine with malfunctioned controls resulting in an incorrect fuel mixture. But considering the rocket was already tilting when that became visible it isn't clear if that happened first or in response to a control input from another malfunction.

If anything this is a testament to KSP's ever-improving realism. Although having a nearly identical failure was probably a one in a million fluke, your craft behaved very nearly the same as theirs did with the exception that you didn't burst into flames midair since KSP doesn't implement that yet.

Edited July 2, 2013 by OdinYggd

[Kryten]

The initial plume of brown smoke is nitrogen tetroxide venting as the engines increase throttle after liftoff, it's normal for Proton launches. More smoke only appears after that after the aerodynamic forces get strong enough to start ripping off parts of the rocket. It looks more like the initial failure was of one of the engine's gimbal systems, or of the actual flight computer.

[togfox]

Harvester once said that random failures won't ever happen in KSP - this way, mission success depends on player skill alone and not on luck. KSP can be frustrating enough without random failures.

However, I do see this a potential plugin idea - sort of like Deadly Re-entry, something to make the game harder and more realistic for the players who want it that way.

Already done by the KGSS. out of date now.

http://forum.kerbalspaceprogram.com/showthread.php/24003-0-17-Nanobots-Awry-Scenario

I think we should have some more posts about struts, ASAS and safety range detonations. We really don't have enough. And we should get more people that actually read the thread - or even the last two or three posts.

If someone has any genuine facts on this as it develops over time then please post them.

[nhnifong]

Bigger question is did the fault originate as a control system malfunction, or was it an engine malfunction that resulted in a loss of control.

That might be indicated at least in part by whether the small roll at the beginning which resulted in a pitch in the wrong direction and several overcorrections was expected given the minor initial perturbation.

I don't suspect that any guidance system worth it's salt would be capable of such a thing.

I'm just betting my imaginary internet credits that the gimbal on engine 3 failed to actuate at launch. (number 0 is on the side with ßрþтþý written on it, numbers increase clockwise looking at the rocket from the bottom, yes I made up a numbering scheme just so I could bet on it)

[Hedd]

They forgot to switch to the flight AI on the launchpad. Brown globe on liftoff is bad mmmkay.

[Kerbal01]

Russians don't use a range safety device: They merely shut off the engines. In this case, policy dictates that they cannot do so for 42 seconds, in order to ensure that the crash is away from the launchpad.

How was this value (of 42 seconds before engine cutoff) decided, experimentation, or random chance?

[OdinYggd]

How was this value (of 42 seconds before engine cutoff) decided, experimentation, or random chance?

Probably a mix of accident statistics and calculation.

Considering the Russians do have slightly more on-pad explosion experience than the Americans do, with the N1 and others like it.

Likely they figured out that if the ship holds together for the all important 42 seconds it will be sufficiently far downrange that the resulting explosion will be unlikely to injure ground support personnel or spectators in the observation areas.

NASA on the other hand usually has to detonate midair so that no large debris come down on populated areas. Although the firebrands can be damaging, at that point most of the risk of further explosion is gone and they can be handled as any ordinary fire.

[EnragedPlatypus]

It's been an accepted fact that more engines = more chances to fail (that is, if the engine was the problem). Is the official summary released already?

I feel like I watched a Russian launch not too long ago where the vehicle (not a Proton) had lost an engine but still managed to get into orbit. I also remember a case of SpaceX losing an engine on one of their launches and continuing fine.

[OdinYggd]

I feel like I watched a Russian launch not too long ago where the vehicle (not a Proton) had lost an engine but still managed to get into orbit. I also remember a case of SpaceX losing an engine on one of their launches and continuing fine.

Engine-out is a design feature not found in all launchers.

SpaceX Falcon 9 is able to, as was the Saturn V itself since more often than not the center engine had to shut down early because of POGO problems.

From the layout of the engines in the Proton-M it probably did not have that capability. The Space Shuttle didn't have it either, if a SSME failed during launch their only chance of escape was to hope the SRBs got them high enough to ditch all boosters and tanks and try to land again.

[Tiberion]

The space shuttle did have single engine out capability, and it in fact USED this in STS-51-F (July 1985) when Challenger lost one of the pumps on the center engine and they performed the Abort To Orbit (ATO) recovery plan. Despite the lower initial orbit, the mission continued as planned.

Challenger even flew again 3 months later (and 3 months after that too, which was the tragic mission where it was lost)

Also, I think Soyuz also has engine-out capability to some extent. I don't know that Proton does though.

[ShachonianX]

Just so you know, this isn't exactly that funny. 3 something billion satellites were destroyed, and toxic fumes from the fuel are settling in cities, killing livestock.

[Galacticruler]

They forgot the ASAS and control surfs.

[Nibb31]

They forgot the ASAS and control surfs.

You know, that wasn't funny the first time. It still isn't when you're the 100th person to pull the same joke.

[refsmmat]

Saturn V did not "more often than not" have the inboard engine fail due to pogo. Once again, that's a feature, not a bug. On both S-1C and S-II inboard engine was scheduled to cut out before the four outboard engines to soften the impact of staging events. On Apollo 6, S-II engines #2 and #3 failed due to pogo complications, neither of them were the inboard engine.engine failures due to pogo. On Apollo 13 the inboard S-II inboard engine did cut out ahead of schedule due to pogo. The other 11 Saturn V launches didn't have any engine failures due to pogo.

[PakledHostage]

Just so you know, this isn't exactly that funny. 3 something billion satellites were destroyed, and toxic fumes from the fuel are settling in cities, killing livestock.

I think it was closer to $200 million for the satellites. And I don't think anyone over about 14 (physical or mental age) thinks this is funny. It is, however, natural curiosity to play "accident investigator" and try to figure out what happened from the evidence available. The silver lining in that cloud of smoke is that launch industry professionals will do their own investigation, and hopefully launch safety will be improved as a result.

[Kryten]

New video, from very close;

Looks like there was definitely one engine not vectoring with the rest.

[Bunsen]

New video, from very close;

Looks like there was definitely one engine not vectoring with the rest.

I see two exhaust plumes on opposite sides of the rocket that don't seem to be vectoring. That might be normal behavior if you're right about this:

Engine gimballing. One plane per engine, I think tangential to the main rocket body.

A yaw error, from the camera's point of view, would require thrust vectoring along a plane perpendicular to the gimbalability (yes that's a word now, dammit) on the leftmost and rightmost engines, so they wouldn't participate. If there's also a pitch error, then those engines' vectoring would be visually inconspicuous from our point of view.

[TouhouTorpedo]

It looks like one engine is gimballing in the wrong direction. If so, then that leads back to some quite entertaining implications of how thats even possible. It does make a lot of sense why the rocket rolled quite strongly after it went off course if one engine is correcting in the other direction.

Be interesting to see what could have caused such strange behavior. I doubt its possible to install the engine backwards, wiring it backwards is a possibility however implausible it sounds. You'd think those sorta things wouldn't happen, but then you also wouldn't think the Genesis probe would have crashed because the parachute accelerometer was on backwards, or that mars climate orbiter would have instead of gain a nice orbit around mars fly straight into it due to using the wrong units. Unfortunately, space programs still have the same unreliable faulty piece of equipment common to every step they did 50 years ago. Us.

EDIT:

wait, rewatched it. Its still strange one engine is behaving differently but I've just noticed something. There doesn't appear to be any differential control going on here. when the rocket is nearing straight up its still thrusting into the turn, something it absolutely should not be doing. When you add time lag from the system in (if a computer says, "turn motor to here" it'll take a while for the motor to respond, especially on something really big) then thats just setting up for some really nasty oscillations.

So on rewatch and reevaluation of what I've seen, I'm not convinced the engine was the cause at all. This was quite certainly a software problem in the controller.

Edited July 3, 2013 by TouhouTorpedo

[Kryten]

I'd say guidance system failure is indeed the most likely option at this point. This was a one-off rocket; a phase-I Proton-M with the RD-276 engines from the phase-II; which would greatly increase the chance of it having some new glitches.

Edited July 3, 2013 by Kryten

[PakledHostage]

Thank you, mods (Rich?), for dredging this thread out of the Junkyard and back into the Science Labs! [dusts off a few coffee grounds and banana peels]

[Kilmeister]

From that angle in Kryten's video it appears starting at :05, just as it clears the smoke cloud, that there is one nozzle thrusting at a different angle than the rest. and it begins to spin around for the next 7 or so seconds before it begins tumbling wildly off course.

[Kryten]

From that angle in Kryten's video it appears starting at :05, just as it clears the smoke cloud, that there is one nozzle thrusting at a different angle than the rest. and it begins to spin around for the next 7 or so seconds before it begins tumbling wildly off course.

But at 0:07 the opposite engine can be seen doing the same thing, as Bunsen said. However, that one engine does appear to have a shorter plume that others, as so;

Maybe we're looking at engine failure after all. That could still be a guidance issue though, those engines can be throttled for increased control authority.

[Kilmeister]

Actually after I've looked at it again. I think the angle differences you can start seeing at 5 seconds onwards might just be reactionary as whatever is causing the rocket to go off course is already having a great effect by that point. I was paying close attention to the flames and not the angle of the rocket. It looks like it is already off course in the first few seconds. By 5 seconds in, that one nozzle is probably just attempting course correction.

This doesn't really rule out the guidance system, but my earlier analysis doesn't seem to apply.