Wednesday, March 21, 2007

Falcon 1 Demo 2 Flight Highlights

By now you've probably heard that Falcon 1 flew yesterday, but did quite achieve orbit. The first stage burn looked beautiful, but it looks like some kind of instability built up during the second stage burn. The aRocket list was abuzz with the term 'coning' to describe a circular resonance around the roll axis of the rocket (nose to tail). I was wondering if this might be due to swirling in the propellant tanks, not unlike how one swirls wine around in a goblet, which would be an unusual case for the engine servo to correct.

At any rate, I caught a few screen caps from the video and have included them, above. Here's a few notes on them:

T +00:13 This is right after Falcon 1 storms off the pad. I just added it 'cause it was cool. Notice the drop of water on the left side of the shot, which I think might be from inside the camera housing. Later on there are some bits of fluff (insulation or LOX-related snow?) that are clearly between the camera and the glass.

T +01:46 Midway through the first stage burn and you can begin to see the exhaust plume beginning to expand as the atmosphere thins out. This gradually increases in size and becomes more and more transparent.

T +02:31 Towards the end of the first stage burn. The exhaust plume is pretty transparent at this point.

T +02:52 First stage separation. Notice that the second stage engine scrapes against the inner-stage just slightly. There was some speculation that this might've been the start of trouble for the second stage burn, but the contact didn't seem all that rough from the camera's perspective.

T +03:13 A few seconds into the second stage burn and a band or ring separates itself from the edge of the rocket bell. Was the purpose of this ring to act as a bumper during first stage sep, or was it more important than that? The Merlin engine is starting to glow red. It will continue to get hotter over the course of the burn.

T +03:17 Shortly after the fairing was jettisoned, it looks like another ring of material might've came off of the engine, or it could've been something related to the fairing. Hard to say.

For nearly two more minutes the burn continues. At first, everything seems quite normal, but in the last 30 to 40 seconds, the 'coning' oscillation builds up, which was a bit disconcerting, but never seemed terribly hideous. The problem was that it was an oscillation that was gradually increasing in amplitude, which was a clue that the control system was having a deuce of a time making it go away.

Additionally, during the second stage burn there were times when quick puffs of black smoke were coming from the engine. I'm wondering if this wasn't related to some kind of combustion instability going on. At around T +05:00, the video feed cut out, so it's hard to speculate happened after that.


This was an amazing flight of a privately funded rocket. The first stage performance was nigh unto flawless. The second stage obviously had some issues. The flight also included one of the most startling false-starts I have ever witnessed: the first stage was actually started twice in one day and over the course of about 30 minutes. I hope SpaceX can figure out what went amiss with the second stage and get back to flying rockets soon.


Blogger bill said...

It's been confirmed that the roll axis oscillation was at the heart of the stage 2 failure. Now the only question is exactly what caused it. The engineering issue coming into this flight has been that SpaceX has had no practical method of fully testing their second stage design until now -- since they've obviously never had the chance to fly it previously. From the point of a non-engineer, this might seem like a real disappointment, but from an engineering perspective, it was a great flight as it answered many engineering questions.

March 21, 2007 9:31 AM  
Blogger bill said...

Since the Falcon-1 payload weighed only 19.5 kg, I'm wondering if this was a case of too light of a satellite and therefore the center of gravity of the upper stage was too far back & being dominated by fuel slosh toward the 2nd half of the burn. This resulted in an unstable servo loop. Two thoughts come to mind -- baffles (to kill the slosh) and ballast (to compensate for the CG).

March 23, 2007 12:42 PM  

Post a Comment

Links to this post:

Create a Link

<< Home