SpaceX Falcon 9 FT - ORBCOMM-2 - Dec. 21, 2015 (Return To Flight) DISCUSSION

[Kabloona]

That's an argument in favor of the idea that they may do a targeting "offset" in which they target coordinates east of the landing pad during the boostback and braking burns, in order to bias the IIP east, then do a final correction with the grid fins during final descent. Only speculation, of course.

OK, what if the boost back burn runs "hot" and then there's a reentry burn, or other failure?  Couldn't this then have a stage plummeting toward someplace like Cocoa or Rockledge, with FTS only able to blast it into many pieces before it hits?

 - Ed Kyle

I guess it depends on what destruct protocol the Range comes up with. If they worry about that scenario, then they can write the rule that says a destruct command will be issued any time the IIP goes west of the pad.

 In your scenario, a hot boostback burn that results in the IIP going west of the pad, a destruct command can be issued immediately, and at that point in the mission the stage is still so far out that the fragments will fall well out to sea, once they hit atmosphere and start tumbling.

[LastStarFighter]

That's an argument in favor of the idea that they may do a targeting "offset" in which they target coordinates east of the landing pad during the boostback and braking burns, in order to bias the IIP east, then do a final correction with the grid fins during final descent. Only speculation, of course.

OK, what if the boost back burn runs "hot" and then there's a reentry burn, or other failure?  Couldn't this then have a stage plummeting toward someplace like Cocoa or Rockledge, with FTS only able to blast it into many pieces before it hits?

 - Ed Kyle

That would be my concern.

I'm curious how far off the coast the target would need to be to keep the entire error ellipse in a safe place. And how much steering capability the stage really has to correct for it.

[alang]

Seems to me that SpaceX has already proved what is necessary - they can hit a target and their flight termination system works.
No operator can guarantee that a returning stage won't crash, surely what is important is to manage the crashes that will happen.

[Coastal Ron]

OK, what if the boost back burn runs "hot" and then there's a reentry burn, or other failure?  Couldn't this then have a stage plummeting toward someplace like Cocoa or Rockledge, with FTS only able to blast it into many pieces before it hits?

No need for FTS if they just shut off the engine - that solves the "many pieces" problem.  And they can do that at any point it deviates from the safe passage corridor.

[cdleonard]

It's worth noting that this is a very light payload.

The last orbcomm flight carried 6 satellites of 172 kg each into orbits ~700km high. This time the payload contains 11 satellites but that still only adds up to ~2 tons. There is also a payload adapter involved but that can't be terribly heavy. This is much less than their advertised 13 tons to LEO.

This flight has a lot margin and I guess this is why they are attempting to land back at the cape. I expect that later flights will be restricted to using a barge for recovery.

[cosmicvoid]

OK, what if the boost back burn runs "hot" and then there's a reentry burn, or other failure?  Couldn't this then have a stage plummeting toward someplace like Cocoa or Rockledge, with FTS only able to blast it into many pieces before it hits?

No need for FTS if they just shut off the engine - that solves the "many pieces" problem.  And they can do that at any point it deviates from the safe passage corridor.

I thought that during the return flight, the F9 is autonomous (like during the outbound leg), so it cannot be commanded to do anything. Only the FTS can effect its path.

[Kabloona]

I thought that during the return flight, the F9 is autonomous (like during the outbound leg), so it cannot be commanded to do anything. Only the FTS can effect its path.

That is correct.

[ChrisWilson68]

OK, what if the boost back burn runs "hot" and then there's a reentry burn, or other failure?  Couldn't this then have a stage plummeting toward someplace like Cocoa or Rockledge, with FTS only able to blast it into many pieces before it hits?

No need for FTS if they just shut off the engine - that solves the "many pieces" problem.  And they can do that at any point it deviates from the safe passage corridor.

I thought that during the return flight, the F9 is autonomous (like during the outbound leg), so it cannot be commanded to do anything. Only the FTS can effect its path.

You're missing the point.  The point is that the avionics on the first stage know exactly where it is and whether it's on the correct path or not.  So there's no need for an outside command.  If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

If the avionics somehow go completely bonkers and fail to turn off the engines in that situation, that's when the FTS is used.  And the pieces will fall harmlessly into the sea.  Until the final few seconds, it will likely never be on a path where the ballistic impact point will be over land.

[baldusi]

Btw, the many pieces following an FTS activation have a lot more aero drag than the stock stage. That's why it would still work.

[ChrisWilson68]

It's worth noting that this is a very light payload.

The last orbcomm flight carried 6 satellites of 172 kg each into orbits ~700km high. This time the payload contains 11 satellites but that still only adds up to ~2 tons. There is also a payload adapter involved but that can't be terribly heavy. This is much less than their advertised 13 tons to LEO.

This flight has a lot margin and I guess this is why they are attempting to land back at the cape. I expect that later flights will be restricted to using a barge for recovery.

You're right that this flight has a very high margin, but with the upgrades to the v1.1FT version they can do well over 13 tons to LEO and still return the first stage to the landing site.

[Kabloona]

If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

That's possible in theory, but have we heard anything that suggests SpaceX has implemented that in their avionics design? I haven't heard anything to that effect. That's what FTS is for.

Just saying unless you know for sure that's how the system is designed, you might want to say "may turn off the engines" instead of "will turn off the engines."

[Lee Jay]

If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

That's possible in theory, but have we heard anything that suggests SpaceX has implemented that in their avionics design? I haven't heard anything to that effect. That's what FTS is for.

Just saying unless you know for sure that's how the system is designed, you might want to say "may turn off the engines" instead of "will turn off the engines."

We saw it actually happen in Texas.

[Jarnis]

If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

That's possible in theory, but have we heard anything that suggests SpaceX has implemented that in their avionics design? I haven't heard anything to that effect. That's what FTS is for.

Just saying unless you know for sure that's how the system is designed, you might want to say "may turn off the engines" instead of "will turn off the engines."

We saw it actually happen in Texas.

Indeed. F9 autonomous FTS seems to be "turn off engines, unzip tanks" which then generally results in rapid mixing of RP1 and LOX, followed by Kaboom.

[ugordan]

We saw it actually happen in Texas.

Yes because there was no other range safety option available. It was done on an experimental vehicle as a means to satisfy FAA safety requirements. What says SpaceX deployed the same logic on orbital missions where a false positive from an autonomous FTS box means dumping tens of millions of $ into the drink?

[Kabloona]

If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

That's possible in theory, but have we heard anything that suggests SpaceX has implemented that in their avionics design? I haven't heard anything to that effect. That's what FTS is for.

Just saying unless you know for sure that's how the system is designed, you might want to say "may turn off the engines" instead of "will turn off the engines."

We saw it actually happen in Texas.

No, what we saw in Texas was FTS activation resulting in engine shutdown followed by destruct.

Chris Wilson is suggesting there is a separate flight computer/avionics logic entirely outside the self-contained FTS system that will turn off the engines *without* activating FTS and subsequent destruct.

[JamesH]

If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

That's possible in theory, but have we heard anything that suggests SpaceX has implemented that in their avionics design? I haven't heard anything to that effect. That's what FTS is for.

Just saying unless you know for sure that's how the system is designed, you might want to say "may turn off the engines" instead of "will turn off the engines."

Seems like a sensible approach though. If by turning off the engine you can dump a single object in to the sea, would that be better than dumping a load of smaller objects over a larger areas, had FTS been activated.

[notsorandom]

Btw, the many pieces following an FTS activation have a lot more aero drag than the stock stage. That's why it would still work.

What about the engines? I don't think there are destroyed as part of the FTS activation. Compared to the rest of the stage they are denser.

[JamesH]

Going back a bit, how far offshore would a barge need to be stationed before you don't need launch range approval to land there?

SpaceX could test the vast majority of flyback and still land a couple or more miles off shore.

[Kabloona]

If the stage leaves its assigned safe flight path, the stage itself will detect it and turn off the engines.

That's possible in theory, but have we heard anything that suggests SpaceX has implemented that in their avionics design? I haven't heard anything to that effect. That's what FTS is for.

Just saying unless you know for sure that's how the system is designed, you might want to say "may turn off the engines" instead of "will turn off the engines."

Seems like a sensible approach though. If by turning off the engine you can dump a single object in to the sea, would that be better than dumping a load of smaller objects over a larger areas, had FTS been activated.

It's not a sensible approach from SpaceX's point of view because it introduces another failure mode in which a false positive will cause unnecessary loss of vehicle, as ugordan pointed out above.

And from the Range's POV, they are always going to want a positive, definite way to render the stage non-propulsive, which means unzipping the tanks. They'll just write the destruct rules in such a way that they're confident the debris won't be a hazard, ie destruct sooner rather than later. The onus will be on SpaceX to keep the stage's IIP in a tight box or suffer the consequences.

The Range is evolving towards autonomous FTS for outbound vehicles, and maybe eventually autonomous FTS will be implemented on returning stages, but for now they're doing it old-school.

[JamesH]

These devices are complex already, with many failure modes, adding something like this is not a huge change. They already have autonomous guidance etc, this is 'simply' a small addition to that. It's not like the software isn't massively tested before flight. 

The system will know to a high degree of accuracy where it is going to land once engines are off. Probably pinpoint if the RCS/grid fins are still functioning. Is that better than blowing the device up and throwing stuff off in random directions? Dunno. It's even possible that it would still be able to do a final deceleration/landing burn and make a soft water landing. Useful? Dunno.

And of course, FTS would still be there if things goes very badly wrong.

It's similar to the principle of an aborting cargo Dragon capsule leaving an exploding stage, had it had the right software to enable a parachute deploy, it could have been 'saved'. And they are going to put in software for that very circumstance.