Launch News (FAILURE) Proton-M/Block-DM-03 launch with triple GLONASS-M, July 2, 2013

[RisingFury]

Right, so with sensors inverted, the rocket computer attempted to stabilize the vehicle in a perfect nose-down trajectory. And it almost succeeded. :blink:

It more looks like some rate gyros had not been installed orthogonal to the others, since there was a spinning out of control.

But it does still not explain the early launch.

We have no way of knowing what the flight computer did without actually knowing the output of the sensors at the time and looking at the software.

 

[Thunder Chicken]

But it does still not explain the early launch.

Accelerometers reported a + 9.81 m/s^2 acceleration vs. - 9.81 m/s^2, and the computer thought it was accelerating upwards at 1 g?

 

[Urwumpe]

Accelerometers reported a + 9.81 m/s^2 acceleration vs. - 9.81 m/s^2, and the computer thought it was accelerating upwards at 1 g?

No, the signal is generated very simple by a contact pin at the base of the rocket. The release of the contact pin is purely mechanic, the pin is about 5 mm long, and follows the rocket for 11 mm more before it gets pulled out by the rising rocket.

Possibly the umbilical was not correctly installed, but that would have been another human failure, and the poor guy that just started his job can't have been everywhere.

 

[N_Molson]

and the poor guy that just started his job can't have been everywhere.

Who knows, maybe they hire 3 rookies for a ridiculous salary and let them build a whole Proton-M alone...

"Uh yeah there are a few wires I'm unsure about but uh yeah, I think it should work !"

 

[Urwumpe]

"Now, the assembly manual says «Bolt part 24321 to part 29922 using bolt D5» but I have no D5 bolt left... maybe another bolt."

 

[N_Molson]

"Hmm, that rusty one is too loose... bah, let's put some glue, nobody will notice."

 

[Artlav]

nobody will notice."

And that's the attitude that kills.
"Checking the ticks, jumping through the hops" as the end goal.
Not "make sure the rocket flies high and true" as the end goal.

Quality control is not a fiend to slip past at any cost, but a friend to double check your work.

Unfortunately, the modern, "reformed", education here is focusing on how to cheat until the numbers appear right, rather than on thinking what it takes to get the job done.

 

[Donamy]

"I thought you meant, my right." :facepalm:

 

[SiberianTiger]

"Use of the force detected"...

http://russianspaceweb.com/proton_glonass49.html#official

[FONT=Verdana, Arial, Helvetica, sans-serif]Agency confirms key facts about the accident

[/FONT] [FONT=Verdana, Arial, Helvetica, sans-serif]On July 18, Roskosmos confirmed all key facts about the accident reported here previously. The agency's statement stressed that the launch vehicle did lift off 0.4 seconds earlier than scheduled, however the following analysis and simulations of the flight had shown that this situation could not had caused the accident. At the same time, the agency confirmed that the launch vehicle had lost stabilization along its course axis, as a result of the improper installation of the DUS angular velocity sensors 180 degrees away from their correct direction. The investigation revealed that three out of six DUS instruments had shown marks of force applied to their docking surfaces and post-accident simulations conducted on mockups of the system had left similar impressions, Roskosmos said.[/FONT]
[FONT=Verdana, Arial, Helvetica, sans-serif]
[/FONT]
[FONT=Verdana, Arial, Helvetica, sans-serif]The agency concluded that existing methods of quality control during ground preparations and testing (of the rocket) according to current design, technical and operational documentation had not allowed to detect the improper installation of DUS sensors and, therefore, the defect had had a manufacturing nature.[/FONT]
[FONT=Verdana, Arial, Helvetica, sans-serif]
[/FONT]
[FONT=Verdana, Arial, Helvetica, sans-serif]The investigation commission had still continued its work to determine the cause of the premature separation of electric interfaces of the launch pad (from the rocket), Roskosmos announced.[/FONT]

So it appears that the sensors were just hammered into the sockets.

It finally should be understood that just boosting QA staff won't assure reliability if the production is done by hands of those who don't give a damn (and have no reason for giving, too) about the final result.

And it's yet difficult to count how much Roscosmos will lose in the nearest years on the cancelled launch contracts and boosted insurance rates.

 

[boogabooga]

I'm surprised such sensors are not designed such that it is impossible to install backwards, akin to pounding a square peg into a round hole.

 

[N_Molson]

In soviet Russia, the sensors install you...

 

[SiberianTiger]

I'm surprised such sensors are not designed such that it is impossible to install backwards, akin to pounding a square peg into a round hole.

A question like that was asked on the NK forum, who are really to blame, the fool workers who installed the sensors a wrong way, or the designers who didn't design them fool-proof. The response was, fools aren't supposed to work in space industry.

 

[Artlav]

I'm surprised such sensors are not designed such that it is impossible to install backwards, akin to pounding a square peg into a round hole.

"Use of the force detected" means quite literally that a square peg was hammered into a round hole.

The program in the head of the installer likely was "put a tick on the checklist to get the wages" not "make sure the rocket flies right".

 

[boogabooga]

http://russianspaceweb.com/proton_glonass49.html#official



So it appears that the sensors were just hammered into the sockets.

It finally should be understood that just boosting QA staff won't assure reliability if the production is done by hands of those who don't give a damn (and have no reason for giving, too) about the final result.

And it's yet difficult to count how much Roscosmos will lose in the nearest years on the cancelled launch contracts and boosted insurance rates.

"Use of the force detected" means quite literally that a square peg was hammered into a round hole.

The program in the head of the installer likely was "put a tick on the checklist to get the wages" not "make sure the rocket flies right".

Perhaps it's time for you gentlemen to get your resumes together and make some calls... :tiphat:

 

[SiberianTiger]

Perhaps it's time for you gentlemen to get your resumes together and make some calls... :tiphat:

Late by 20 years, after taking a wrong education line. They'll only consider me for a backyard cleaning position.

 

[N_Molson]

-"Where do you work ?"
-"Oh, in a russian aerospace company."
-"Hey but that's cool. What's your speciality, rocket engines, satellites, aerodynamics ?" :thumbup:
-"Backyard cleaning." :dry:

 

[Cosmic Penguin]

Slow motion of the crash, in HD!

Bumping this thread because I have just seen the most amazing video of this crash ever, courtesy of Nicholas Pillet of kosmonavtika.com via an unnamed source! :blink:

 

[RisingFury]

That image looks like something straight out of KSP.

 

[N_Molson]

Don't breath the smoke. Just an advice.

 

[RisingFury]

Don't breath the smoke. Just an advice.

If the rocket gets close enough to you that you can inhale the smoke, I think you won't have time to die of poison gas inhalation.