Wow. That’s bad design and bad QC.
a) You shouldn’t be able to install an angular-rate (or any kind of attitude) sensor upside down.
b) The vehicle should be able to sense that it has an upside-down sensor and automatically abort.
It did look like it was having guidance problems in the video, and that would sure explain why.
By July 9, it is transpired that investigators sifting through the wreckage of the doomed rocket had found critical angular velocity sensors, DUS, installed upside down. Each of those sensors had an arrow that was suppose to point toward the top of the vehicle, however multiple sensors on the failed rocket were pointing downward instead.
We’ve all heard of Murphy’s Law but many don’t understand what it really means. While there are many variations on Murphy’s Law, the original one was “Whatever can go wrong, will go wrong.” Another way of saying that is that “If something can be done wrong, it will be done wrong.”
From that, we need to design things so they can’t be done wrong. Many have done just that. Try and plug in a USB cable wrong. You can’t because it was designed to prevent it. And yet, we still have multi-million dollar aerospace systems being designed and built where things (such as installing those sensors) can be done wrong. A few examples:
June 1991 – a miswired flight control computer caused the crash of the fifth V-22 Osprey on its first flight.
There was an incident on an Airbus airliner several years ago (can’t find the link at the moment) where the pilot’s control stick was miswired. That really screws with a pilot’s mind. Fortunately, the copilot’s stick was wired properly and he took over, narrowly avoiding a crash on takeoff.
In 2004, the Genesis sample return system failed to deploy its parachute and crashed into the desert because the G-switches were all installed backwards.
These are just three that quickly come to mind and I’m sure there are more examples. Add the Proton to the list.
Damnit, design things so they can’t be installed backwards and these kind of problems won’t happen. Because if it can be done wrong, it will be done wrong sooner or later.
During the Falklands War an Argentinian submarine never scored a hit with its wire-guided torpedoes because during a refit the control joystick’s left/right wires were swapped.
Chuck Yeager related that a serious of mysterious and fatal aileron lockups on the F-86 were traced back to an assembly line employee who was installing a bolt opposite the design orientation, insisting that nuts always went on the bottom.
Of course, until we make this a firing offense when you know better than to do it wrong (both George’s last example, and supposedly the Proton failure was traced back to a single employee who knew they were supposed to be installed the other way) this will keep happening.
The guy was just lucky Stalin is no longer in charge. If he was he would have been denounced as a wrecker and promptly shot.
Try and plug in a USB cable wrong. You can’t because it was designed to prevent it.
I always have it the wrong way around when I first try to plug it in.
Look at the cable. All but the cheapest ones are marked. The side with the usb logo is opposite the side with the contacts; in most devices it’ll point up when plugged in. (For example, front-panel USB sockets on computers ALWAYS align this way.)
This isn’t common knowledge, for some reason.
I trashed the Ethernet port on one laptop because although a USB cable can’t plug into a USB port upside down, it can be shoved far enough into an Ethernet RJ45 jack to damage its exposed and tensioned wires when you’re reaching around the back of the PC and going by feel.
IIRC, one of the early Apollo launches had some wiring going to the wrong engine and they fixed the issue by making the wiring shorter so they could only reach the intended engine.
Try and plug in a USB cable wrong. You can’t because it was designed to prevent it.
You’re just not pushing hard enough.
In fact the occasion for the coining of Murphy’s Law was a technician installing sensors the wrong way up, because the way they were designed meant that they would fit both ways……
Sounds like a perfect match to this case…
0 failures can be had at 1/0 cost
An article in a Russian newspaper (linked to by NasaSpaceflight.com) said the sensors had arrows on them to prevent this from happening. Unfortunately a new technician installed the sensors with the arrow pointing downward, not upward. New Russian proverb: if sensor arrow points downward, you will not go to space today. ( hat tip xkcd, http://xkcd.com/1133/)
You have to admit that the arrows worked. The arrows pointed down and the rocket most definitely went down.
Well, fixes are simple.
Visually verify that the sensors and all other equipment is mounted correctly.
Design all equipment mounts so they can’t be mounted incorrectly, for example, by not having mounting screw locations that have a rotational symmetry.
Test all the accelerometers, position sensors, and rate sensors by powering them up while the rocket is being moved or erected, because even if the sensor assembly is mounted correctly, the sensor itself could be installed incorrectly inside the housing.
It is also not very difficult to invert, scale, offset, and rotate both digital and analog signals so that they make logical sense based on other sensors and what the vehicle must be doing. That would be similar to a voting system, yet instead of just ignoring an errant sensor, the flight computer would divine its true intent, like reading a Florida chad.
You can always measure Earth rate on rate sensors, and the normal force of the surface reacting against gravity on accelerometers.
I think you’re all being far too critical, and should take a cue from our ever more egalitarian schools: The rocket didn’t fail, it just expressed itself via its components going in individualistic directions.
But if we make it so that rockets cannot be built incorrectly, how will we make any new cool videos of rockets going off course?