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

[LouScheffer]

This is a very light payload for this rocket.  If they are already worried about first stage return-to-launch-site margins, how is this design supposed to work with "real" payloads?

 - Ed Kyle

I could imagine the rule for the first stage, for each mission, says "stop when you have X percent of fuel left", where X should be (barely) enough for landing.  For any payload, light or heavy, this gets maximum performance from the first stage, which maximizes mission success likelihood (second stages have been known to underperform, see Delta-IV).  The X percent would need to be measured by volume, since as far as I know there is no way to measure mass.  The good thing about such as rule is that any extra margin from a light payload goes towards mission success, not for an easier landing.  This seems the right priority to me.

This scenario, which is just a guess, explains both comments.  Orbcom is willing to wait since cooler, denser fuel means more delta-V total and hence better odds for their mission in case of under-performance.  SpaceX likes denser fuel, too, since it gives them more delta-V for landing (plus the winds are expected to be calmer tomorrow)

[Kim Keller]

This is a very light payload for this rocket.  If they are already worried about first stage return-to-launch-site margins, how is this design supposed to work with "real" payloads?

 - Ed Kyle

I could imagine the rule for the first stage, for each mission, says "stop when you have X percent of fuel left", where X should be (barely) enough for landing.

I consider that unlikely. It would place more priority on getting the S1 back than getting the payload to the proper orbit. Typically - on liquid fuel boosters - the first stage burns until certain trajectory requirements are met, like speed and altitude along a given azimuth. That's why you'll see BECO occurring at different times from mission to mission. If the S1 underperforms, it then will have to use some of its landing fuel margin in order to get to a staging point where the S2 can get its payload where it needs to be.

[Bubbinski]

I wonder if there's something in the test fire data they want to look at more closely? It seems to me that reading between the lines from the 2 announcements that landing isn't the only consideration.

[llanitedave]

I know the real reason for the delay.  It's because I have to get off the computer and out the door at 5:30 p.m. tomorrow evening in my time zone, or 3 minutes before launch.  They're trying to keep me away!  I can't check back in until another 2 or 3 hours later.

Dagnabit.  I take these things personally.

[LouScheffer]

This is a very light payload for this rocket.  If they are already worried about first stage return-to-launch-site margins, how is this design supposed to work with "real" payloads?

 - Ed Kyle

I could imagine the rule for the first stage, for each mission, says "stop when you have X percent of fuel left", where X should be (barely) enough for landing.

I consider that unlikely. It would place more priority on getting the S1 back than getting the payload to the proper orbit. Typically - on liquid fuel boosters - the first stage burns until certain trajectory requirements are met, like speed and altitude along a given azimuth. That's why you'll see BECO occurring at different times from mission to mission. If the S1 underperforms, it then will have to use some of its landing fuel margin in order to get to a staging point where the S2 can get its payload where it needs to be.

I agree with you 100% that the first priority should be getting to the payload to orbit, and if that leaves not enough fuel for landing, so be it.  The two strategies are almost equivalent if you set the BECO speed to that expected for the given payload mass and X% of fuel left.   This too gives all excess performance to the mission and not the landing.  (There is a chance the BECO strategy prevents a landing for a small stage-one performance shortfall that the second stage could easily make up on a light mission.  The percentage fuel cutoff strategy could allow a mission failure given a large second stage underperformance, which would have succeeded had a landing not been attempted.   This is true of any reusable rocket mission, though.)

It seems to me you'd always want to land with minimum residuals, to reduce load on the landing gear and minimize damage in case of an accident.  So it makes sense to not overkill the landing fuel even if you have lots of margin.

[parhaml]

If the S1 underperforms, it then will have to use some of its landing fuel margin in order to get to a staging point where the S2 can get its payload where it needs to be.

Newb question . . .

In the case of (insert vehicle here) S1 underperformance, would flight software use the landing-margin to put S2 within the middle of it's envelope? Bottom 10%, 25% . . . .  Thanks in advance!

[QuantumG]

I agree with you 100% that the first priority should be getting to the payload to orbit, and if that leaves not enough fuel for landing, so be it.

For now.. but eventually the payloads will be worth less than the rocket (otherwise, what's the point of reusability?) and at that point it'll make more sense to save the rocket than the payload.

[speedevil]

This may be one more of an endless series of repeated answers but the ASDS was supposedly able to hold its position to 3 meters. 
This is less than the diameter of the rocket and not significant on the scale of the ~53 meter wide ASDS.
Another ten feet of deck wouldn't have enabled either of the previous landing attempt to succeed.

Weeeel.
To land properly you need to get down at the right vertical, horizontal speeds, in the right horizontal and vertical position.
These are not unrelated problems.
If you relax any constraint, others get easier.
Getting it upright at 0 horizontal speed and 0 vertical velocity is moderately eased - especially if you have an imperfectly tuned control loop - if you add flex to the absolute position.

[Robotbeat]

Wonder how Orbcomm feels about this...

I've got a sneaky feeling they got a very, very cheap ride (relatively speaking) to go as RTF and ride along with what is a validation flight of the upgraded F9 and a landing attempt.

Yup. They probably paid around $10 million. Originally supposed to be a Falcon 1e or something.

[Kim Keller]

If the S1 underperforms, it then will have to use some of its landing fuel margin in order to get to a staging point where the S2 can get its payload where it needs to be.

Newb question . . .

In the case of (insert vehicle here) S1 underperformance, would flight software use the landing-margin to put S2 within the middle of it's envelope? Bottom 10%, 25% . . . .  Thanks in advance!

Yes.

[clongton]

... they passed the data and the LRR, then not happy again. Bit confusing

They didn't say they "weren't happy". What I get from the tweet is that they like the landing odds tomorrow better - that's all. I assume that is because the wind gusts at LZ-1 will be half the strength v.s. tonight's wind gusts. Unless there is a compelling reason to launch tonight, I'd make the same choice, especially because of the importance of this landing attempt. Lighter wind gusts means better control authority for the landing attempt. It's a good call.

[Kabloona]

Wonder how Orbcomm feels about this...

I've got a sneaky feeling they got a very, very cheap ride (relatively speaking) to go as RTF and ride along with what is a validation flight of the upgraded F9 and a landing attempt.

Yup. They probably paid around $10 million. Originally supposed to be a Falcon 1e or something.

SFN is reporting they paid $42.6M for the two flights. Still a bargain.

[CyndyC]

From what I read or heard somewhere back in July reports on CRS-7, Elon gives out his email address at SpaceX, and encourages anyone & everyone to email him with any concerns no matter how small or how last-minute prior to launches. This could have been something that came out of left field from our perspectives, maybe even from a post a SpaceXer read on NSF (LOX thread comes to mind), hence confusing, and they don't want to look silly to the public for what might look like sweating the small stuff, and "Elon being paranoid again," in Elon's own words. Of course I could be wrong, but right now my guess is whatever else came up, the better landing odds were thrown in to appease Elon, and Elon shared that appeasement with the public. Does that sound like Elon Musk? I haven't read the biography.

[WHAP]

Having a little trouble reconciling Musk's tweet with Orbcomm's.  Increased odds of landing vs. looking more closely at static fire data aren't really the same thing.  A bit of obfuscation, IMO.

[faramund]

Surely they don't mean that there is a 10% greater chance the payload will tomorrow be destroyed, surely its there's a 10% greater chance that weather will delay the launch tomorrow.

So they're saying, ok.. we want a greater chance of stage return, and if that delays the launch, then so be it.

That doesn't seem like an irrational choice.

[Lars-J]

Surely they don't mean that there is a 10% greater chance the payload will tomorrow be destroyed, surely its there's a 10% greater chance that weather will delay the launch tomorrow.

Indeed.

[LouScheffer]

Having a little trouble reconciling Musk's tweet with Orbcomm's.  Increased odds of landing vs. looking more closely at static fire data aren't really the same thing.  A bit of obfuscation, IMO.

Orbcomm said:
   - "an additional day prior to launch will allow for more analysis"  this is certainly true, whether or not it's a reason.
   - "and time to further chill the liquid oxygen in preparation for launch." (so more delta-v in case of problems)
Musk said:
   - "Monte Carlo runs show tmrw night has a 10% higher chance of a good landing."  Probably lower winds, but maybe more delta-V available with lower temp fuel.

And no-one said:
   - Gives launch team a little time to rest after a few hard days, and reflect if launch fever might be operating.

I see no contradiction here - the combination of all these reasons made Monday look better than Sunday.   Then each tweeted what they thought was the most compelling reason.

[tobi453]

Percent or percentage point. Yeah, math is difficult...

[Jim]

Having a little trouble reconciling Musk's tweet with Orbcomm's.  Increased odds of landing vs. looking more closely at static fire data aren't really the same thing.  A bit of obfuscation, IMO.

Orbcomm said:
   - "an additional day prior to launch will allow for more analysis"  this is certainly true, whether or not it's a reason.
   - "and time to further chill the liquid oxygen in preparation for launch." (so more delta-v in case of problems)
Musk said:
   - "Monte Carlo runs show tmrw night has a 10% higher chance of a good landing."  Probably lower winds, but maybe more delta-V available with lower temp fuel.

And no-one said:
   - Gives launch team a little time to rest after a few hard days, and reflect if launch fever might be operating.

I see no contradiction here - the combination of all these reasons made Monday look better than Sunday.   Then each tweeted what they thought was the most compelling reason.

The reason with the  customer's buyin would be the most compelling one.

[Kabloona]

   - Gives launch team a little time to rest after a few hard days, and reflect if launch fever might be operating.

Indeed. The multiple rationales given makes one wonder if this isn't the real reason.

The reason with the  customer's buyin would be the most compelling one.

As in, maximize probability of mission success.