FAILURE: Firefly Alpha 1st flight - Vandenberg SLC-2W - 3 Sep 2021 (01:59 UTC)

[JEF_300]

When the vehicle was destroyed, it was (roughly) over the beach, and heading out to sea. Heavy debris still managed to get back near the pad, and could conceivably have damaged expensive infrastructure. Also, light debris, largely harmless, managed to ride upper level winds out to where it could fall on civilians and their homes.

Yes, a scrap of carbon fiber bumping a kid on the head at the playground probably won't even injure them. That doesn't mean it's acceptable.

Yes, heavy debris falling near the pad would happen during an early RUD, and there would be no way to stop it. But this wasn't an early RUD, this was an intentional detonation of the vehicle more than 2 minutes into flight, so it was probably avoidable.

I don't think anyone is making a bigger deal out of this than it is. Neither of these are a crises, and neither are problems that need to be solved.
But they are probably solvable, so why not solve them?

[Helodriver]

At the risk of bringing "those guys" into the discussion, this might be a good time to remind everyone about the Dragon inflight abort test. The F9 was not commanded to unzip; but to simply shut down its engines right around the max-q regime. The vehicle started tumbling, broke apart, and exploded anyway. As expected.

Well, except for the second stage, which smashed back into the ocean intact and then exploded.

Now, had F9 followed a firefly-like 'doomed-from-the-start' trajectory rather than a nominal trajectory, the engine compartment and some tufts of carbon fiber are the least of your worries. You could easily have had a fully fueled second stage come crashing down somewhere on your facilities.

In my eyes, this is very simple. There were no casualties, nor loss of property that I know of. Thus, the RSO did their job competently and properly. If the range makes a process change of some kind, I'll consider it an improvement for the future; not indicative of something they should have done differently in the past. YMMV.

Not to beat a dead horse, but if you watch the replays of the F9 in flight abort test, the rocket does not tumble before the fireball. What it does do is vent a lot of pressure before deflagration.

Rocket 1st stages are tough, so long as they retain flight pressure. Astra 3.1 fell to the ground intact minus its shredded payload fairing after engine shutdown and exploded on impact.  1986's Delta GOES-G stayed intact, again minus the fairing, despite multiple somersaults through its own exhaust just after Max-Q until popped by the range.  Firefly if not command destructed, would likely have similarly remained whole save for the fairing. Given its straight up trajectory, it would also have hit land, whole or not.

[meekGee]

...
In my eyes, this is very simple. There were no casualties, nor loss of property that I know of. Thus, the RSO did their job competently and properly
...

Nobody/nothing got hurt so there wasn't a problem?

That's...  so wrong.

[JEF_300]

In my eyes, this is very simple. There were no casualties, nor loss of property that I know of. Thus, the RSO did their job competently and properly. If the range makes a process change of some kind, I'll consider it an improvement for the future; not indicative of something they should have done differently in the past. YMMV.

I don't think anyone is saying that there was a mistake or error, just that this is a problem, and that things should be done differently in the future. The Range Safety Officer absolutely did everything right.

[Lars-J]

The most noteworthy part, to me, is that none of this would have happened if they had merely shut the engines off instead of blowing the thing up.

Wrong on both accounts.  The vehicle would have still broken up and exploded send parts everywhere.
Can't say it would still be stable after max q.  You have no data to support that claim.

I'm saying that they should have terminated the flight by shutting down the engines, not blowing the vehicle to smithereens. When the range decided to terminate the flight, the vehicle was at-or-just-past max-Q, and was already two tumbles into a spin, with three engines still on, and it hadn't broken up! What in the world makes you think it would encounter a situation after all of that, without the engines on, where it would be under more stress, and thus break up?

I guarantee you that an intact vehicle falling with a significant amount of propellant remaining would cause a LOT more damage than the “blown to smithereens” pieces. As it was, only the engine section appeared to remain large enough to cause significant damage. Can you imagine the whole vehicle?

And the assertion that the vehicle would have fallen into the ocean if unexploded is an one based on very shaky (or no) evidence. The rocket was severely underperforming and and had not gained any significant downrange velocity. And the engine section was not ejected back towards the pad by the explosion, instead going sideways/forward and STILL  landed on land, further disproving a water impact.

The FTS did its job and minimized damage.

[JEF_300]

I'm saying that they should have terminated the flight by shutting down the engines, not blowing the vehicle to smithereens. When the range decided to terminate the flight, the vehicle was at-or-just-past max-Q, and was already two tumbles into a spin, with three engines still on, and it hadn't broken up! What in the world makes you think it would encounter a situation after all of that, without the engines on, where it would be under more stress, and thus break up?

I guarantee you that an intact vehicle falling with a significant amount of propellant remaining would cause a LOT more damage than the “blown to smithereens” pieces. As it was, only the engine section appeared to remain large enough to cause significant damage. Can you imagine the whole vehicle?

And the assertion that the vehicle would have fallen into the ocean if unexploded is an one based on very shaky (or no) evidence. The rocket was severely underperforming and and had not gained any significant downrange velocity. And the engine section was not ejected back towards the pad by the explosion, instead going sideways/forward and STILL  landed on land, further disproving a water impact.

The FTS did its job and minimized damage.

Let's just totally set aside the "would it have made it to the sea" thing for a moment. That's not the really point here, at least not to me.
I, and I'm sure most of us, agree that if it was going to land near the pad, detonation was the obvious choice.
The argument I actually care about making is; if it was going to crash at sea, not detonating would have been better.

[cscott]

The argument I actually care about making is; if it was going to crash at sea, not detonating would have been better.

But that's ridiculous. You can't know with certainty that it will land at sea at the time the FTS decision was made. With an intact vehicle, all sorts of additional events could take place on its way to the ground. Heck, confused software could make the engines relight! Not to mention winds, directed venting of gases, etc etc.

Explosive FTS is designed to shred the vehicle into small nonfunctional pieces and ensure the resulting cloud of debris can be modeled accurately using statistical methods.

[chopsticks]

Sometimes I get the impression that some people here are actively looking for ways to find fault with any decision anyone (or company) makes

[kessdawg]

Just a layman question here: Do the range safety officers track the instantaneous impact point during flight? Do some/any rockets have different flight termination modes based on where the impact point is?

[JEF_300]

The argument I actually care about making is; if it was going to crash at sea, not detonating would have been better.

But that's ridiculous. You can't know with certainty that it will land at sea at the time the FTS decision was made.

Why not? I mean sure, you'll have to have some margins on that calculation, but it's hardly a difficult math problem to have a computer run real-time, even accounting for things like wind. It's just a ballistic trajectory with some modifications. And for those modification, you don't need to actually calculate all the complex aerodynamic effects; you can just assume the worst (assume max surface area into the wind if wind is headed inland, etc.), and move on.

If I had a week to work on it, I'm sure I could write out a program accurate to 300-500 meters or so. The DoD or NASA ought to be able to do better. We know for a fact that SpaceX can do better, because they nail their drone ship landings all the time, and the math involved is largely the same, though more complicated there because they're trying to actually fly, and are trying to be accurate rather than assuming the worst.

With an intact vehicle, all sorts of additional events could take place on its way to the ground. Heck, confused software could make the engines relight! Not to mention winds, directed venting of gases, etc etc.

Ok, now who's being ridiculous? Of the issues of concern on the Firefly Alpha, software starting up the Tap-Off cycle Reaver engines, mid-air, after a non-explosive FTS has been activated, is among the least likely.

But sure, let's say that happens. Solution; if the vehicle suddenly and dramatically changes course on the way down (or it's point of landing moves back onto land because of venting gas or something), blow it up then. I'm sure they don't stop tracking it on radar the moment they terminate the flight, so they would know.

[Comga]

The most noteworthy part, to me, is that none of this would have happened if they had merely shut the engines off instead of blowing the thing up.

Wrong on both accounts.  The vehicle would have still broken up and exploded send parts everywhere.
Can't say it would still be stable after max q.  You have no data to support that claim.

I'm saying that they should have terminated the flight by shutting down the engines, not blowing the vehicle to smithereens. When the range decided to terminate the flight, the vehicle was at-or-just-past max-Q, and was already two tumbles into a spin, with three engines still on, and it hadn't broken up! What in the world makes you think it would encounter a situation after all of that, without the engines on, where it would be under more stress, and thus break up?

I guarantee you that an intact vehicle falling with a significant amount of propellant remaining would cause a LOT more damage than the “blown to smithereens” pieces. As it was, only the engine section appeared to remain large enough to cause significant damage. Can you imagine the whole vehicle?

And the assertion that the vehicle would have fallen into the ocean if unexploded is an one based on very shaky (or no) evidence. The rocket was severely underperforming and and had not gained any significant downrange velocity. And the engine section was not ejected back towards the pad by the explosion, instead going sideways/forward and STILL  landed on land, further disproving a water impact.

The FTS did its job and minimized damage.

I respectfully disagree.
LC-2 is less than 1000 meters from the shoreline to the west, less than 2000 meters heading straight south.  Probably closer to the former along the trajectory.
Approaching or passing supersonic and Max Q it was going at least 1000 meters per second.
If the trajectory is pitched beyond a few degrees its  moving laterally at over 100 m/s. It probably crossed the coastline well before the “event” and had significant momentum away from the pad.

If it had remained intact, it would have been much less effected by the winds, and flown closer to a parabolic arc that would end in the ocean.

That said, this is not conclusively show that the range safety should have done anything different  for an unguided rocket that had been flying briefly in random directions under power.


Added:  what JEF_300 said

[trimeta]

If it had remained intact, it would have been much less effected by the winds, and flown closer to a parabolic arc that would end in the ocean.

Were the engines affected by the winds? I have a hard time believing that the spin was enough to fling the engines backwards towards land (when they'd otherwise make it to the water) or that their descent was substantively affected by wind.

[kdhilliard]

I find it hard to believe that, had the flight been terminated via engine shutdown, the full stack would have made it back to land ... just as I find it hard to believe that the engine section managed to do so!  But I can't argue with the photographic evidence, and if the engines made it, I suspect the full stack could have as well.

Wherever the full stack would have landed, the destruct avoided a several ton RP-1 spill.

Were the engines affected by the winds?

Sure, but no more than an intact, tumbling stack would have been, right?

MECO was to be at T+163 s, so with the engine shutdown at T+15 s and the anomaly at T+150 s, we can roughly calculate that (15+135*3/4)/163 = 71% of the S1 propellant was burned.  (Perhaps a bit more, as the three functioning engines presumably didn't throttle down as planned at ~ T+57 s.)  So while S2 itself would have still been full and dense, the complete stack should have been less dense than the engine section.

... I have a hard time believing that the spin was enough to fling the engines backwards towards land (when they'd otherwise make it to the water) or that their descent was substantively affected by wind.

The earlier estimate of 1 rev/sec seems way too high.  Counting frames for the last 90° of rotation suggest a rotation rate of 4*(51/30) = 6.8 seconds per revolution.  That drops the 100 m/s tangential speed approximation down to about 15 m/s.  And as was mentioned earlier, the engine section does not appear to have been ejected back toward the pad.

But it did fall from 50,000 feet, and fell for nearly 3 minutes (from 2:41 to 5:38 in Keavon Chamber's video).  Do we know how strong the upper level winds were?

[meekGee]

If it had remained intact, it would have been much less effected by the winds, and flown closer to a parabolic arc that would end in the ocean.

Were the engines affected by the winds? I have a hard time believing that the spin was enough to fling the engines backwards towards land (when they'd otherwise make it to the water) or that their descent was substantively affected by wind.

I did a ballpark estimate upthread. I was surprised by the result, but the spin was really rapid.

That's my only comment here - I think FTS should have activated way before the spin was basically tearing it apart.

[Lars-J]

I'm saying that they should have terminated the flight by shutting down the engines, not blowing the vehicle to smithereens. When the range decided to terminate the flight, the vehicle was at-or-just-past max-Q, and was already two tumbles into a spin, with three engines still on, and it hadn't broken up! What in the world makes you think it would encounter a situation after all of that, without the engines on, where it would be under more stress, and thus break up?

I guarantee you that an intact vehicle falling with a significant amount of propellant remaining would cause a LOT more damage than the “blown to smithereens” pieces. As it was, only the engine section appeared to remain large enough to cause significant damage. Can you imagine the whole vehicle?

And the assertion that the vehicle would have fallen into the ocean if unexploded is an one based on very shaky (or no) evidence. The rocket was severely underperforming and and had not gained any significant downrange velocity. And the engine section was not ejected back towards the pad by the explosion, instead going sideways/forward and STILL  landed on land, further disproving a water impact.

The FTS did its job and minimized damage.

I respectfully disagree.
LC-2 is less than 1000 meters from the shoreline to the west, less than 2000 meters heading straight south.  Probably closer to the former along the trajectory.
Approaching or passing supersonic and Max Q it was going at least 1000 meters per second.

I think you are confused between m/s and km/hr. 1000m/s (3400km/hr) is closer to Mach 3 at that altitude (and it was not very high). It was doing 300-400m/s as it started tumbling.

If the trajectory is pitched beyond a few degrees its  moving laterally at over 100 m/s. It probably crossed the coastline well before the “event” and had significant momentum away from the pad.

If it had remained intact, it would have been much less effected by the winds, and flown closer to a parabolic arc that would end in the ocean.

There is nothing to support that assertion. As I wrote above, the heavy engine section was ejected out of the explosion away from the pad. It would not have been much affected by wind, and *still* landed on land.

[Comga]

You are correct, Lars-J.
Sloppy approximation on my part.
Much lower velocity.
But the rocket SHOULD have been arcing over and heading offshore.

We are, though, in agreement that HOW the engine section came to fall not only on land but farther inland than the launch pad is hard to comprehend.
I'm still trying to make sense of it.

[Yggdrasill]

One aspect that I haven't seen mentioned is that the engines might still be providing thrust even as the vehicle is disintegrating.

How much could that thrust accelerate the engine section in say 0.5 seconds once it's not encumbered with the mass of the upper stage/payload, and most of the mass of the first stage tanks and propellant? I would think it's not trivial.

[Yggdrasill]

One aspect that I haven't seen mentioned is that the engines might still be providing thrust even as the vehicle is disintegrating.

How much could that thrust accelerate the engine section in say 0.5 seconds once it's not encumbered with the mass of the upper stage/payload, and most of the mass of the first stage tanks and propellant? I would think it's not trivial.

Okay, I was curious, so I looked a bit at the numbers. If we say the supplied thrust from 3x reaver was around 50 tons, and the effective mass of the engine section with some piping and propellant was 5 tons, that's a T/W of 10.

In 0.5 seconds you could get as much as 50 m/s lateral velocity imparted on the engine section.

In that time you would burn through around 90 kg of propellant. Which doesn't seem too implausible for remainders in pipes and such.

[su27k]

Reusability on the horizon for small satellite launch providers

The debris model for Alpha is being redone as a result of some pieces of the terminated rocket floating downwind to populated areas. A slight change to the mission profile is being considered to mitigate these concerns. The second flight of Alpha will be another demonstration flight with deployable educational payloads onboard.