Falcon Heavy Separation Method

[spacenut]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in. 

[Lars-J]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Yes. And was this in response to a specific point?

[hkultala]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Not 1-2 minutes for launches with center core RTLS.

One minute more flying time for center core means about 100km more distance for the core before the boostback burn can begin, meaning about 100km longer way back home, requiring MUCH more fuel for the boostback burn.

[mme]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Yes. And was this in response to a specific point?

Yes, if the core throttles down after liftoff then the side booster separation is likely earlier than single stick separation. So lower and slower.

[mme]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Not 1-2 minutes for launches with center core RTLS.

One minute more flying time for center core means something like 100km more distance for the core before the boostback burn can begin, meaning about 100km longer way back home, requiring MUCH more fuel for the boostback burn.

Ignoring the SpaceX promotional video, I don't think all 3 cores RTLS is common.

[hkultala]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Yes. And was this in response to a specific point?

Yes, if the core throttles down after liftoff then the side booster separation is likely earlier than single stack separation. So lower and slower.

FH has so much higher T/W that it would have to throttle very deeply to have effective average T/W worse than F9 and stage lower and slower.


And that deep throttling makes absolutely no sense with center core RTLS.

The center core will typically throttle mildly, and the T/W of FH will still be higher than T/W of F9, so that it will stage higher.

[Lars-J]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Yes. And was this in response to a specific point?

Yes, if the core throttles down after liftoff then the side booster separation is likely earlier than single stick separation. So lower and slower.

Yes, but it is still accelerating a much lighter relative mass - FH's larger payload is dwarfed by the mass of the fully loaded upper stage - in the 110t range. 

For RTLS:
- F9, max upper stage + payload mass for F9: ~120t
- FH (boosters RTLS, center ASDS), max upper stage + payload mass: ~150t?


Even at liftoff, an F9 has a T/W ratio of ~1.33... FH will have the same, even if it lifts off with the center core at 50% thrust! So this should show you that it will go faster/higher before booster staging.

[hkultala]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Not 1-2 minutes for launches with center core RTLS.

One minute more flying time for center core means something like 100km more distance for the core before the boostback burn can begin, meaning about 100km longer way back home, requiring MUCH more fuel for the boostback burn.

Ignoring the SpaceX promotional video, I don't think all 3 cores RTLS is common.

3 core RTLS is the cheapest recovery option.

And they will use the cheapest recovery option the mission allows.

Most FH missions will not be launching 16 tonne satellites to GTO, or 14 tonne payloads towards mars, or 40 tonne station parts to LEO. There just are not much this kind of missions.

They will be launching 6-8 tonne satellites to GTO. And maybe also 20-tonne spy satellites to LEO.
The ones F9 cannot do while recovering the first stage.

And 3-core RTLS is enough for these. So 3-core RTLS will be the MOST COMMON recovery method for FH.

[envy887]

I thought after liftoff, the core engines would throttle down to conserve fuel while the side boosters would throttle up to maximum.  Then at separation, the core would throttle up while the side boosters separate and fly away from the core.  The core would continue to burn another 1-2 minutes before second stage kicked in.

Not 1-2 minutes for launches with center core RTLS.

One minute more flying time for center core means something like 100km more distance for the core before the boostback burn can begin, meaning about 100km longer way back home, requiring MUCH more fuel for the boostback burn.

Ignoring the SpaceX promotional video, I don't think all 3 cores RTLS is common.

OneSpeed's sim shows 34 tonnes to LEO which is 14 tonnes to GTO with the center core landing downrange. My calculations using the simulated mass and velocity at staging (205.2 tonnes and 3248 m/s) indicate that for the center core to RTLS it will have to stage 870 m/s earlier, lowering the payload to LEO to about 25.5 tonnes and the payload to GTO to about 9.5 tonnes.

Virtually all GTO payloads are less than 9.5 tonnes, so this seems like a feasible option should SpaceX want to use it. However, it might be much easier on the center booster to use that performance margin for a longer entry retroburn to reduce entry heating.

[mme]

...
Yes, if the core throttles down after liftoff then the side booster separation is likely earlier than single stick separation. So lower and slower.

Yes, but it is still accelerating a much lighter relative mass - FH's larger payload is dwarfed by the mass of the fully loaded upper stage - in the 110t range. 

For RTLS:
- F9, max upper stage + payload mass for F9: ~120t
- FH (boosters RTLS, center ASDS), max upper stage + payload mass: ~150t?


Even at liftoff, an F9 has a T/W ratio of ~1.33... FH will have the same, even if it lifts off with the center core at 50% thrust! So this should show you that it will go faster/higher before booster staging.

...

3 core RTLS is the cheapest recovery option.

And they will use the cheapest recovery option the mission allows.

Most FH missions will not be launching 16 tonne satellites to GTO, or 14 tonne payloads towards mars, or 40 tonne station parts to LEO. There just are not much this kind of missions.

They will be launching 6-8 tonne satellites to GTO. And maybe also 20-tonne spy satellites to LEO.
The ones F9 cannot do while recovering the first stage.

And 3-core RTLS is enough for these. So 3-core RTLS will be the MOST COMMON recovery method for FH.

These are all excellent points and I can't wait to see this thing fly.

[Jcc]

Of course to state the obvious, there are no signs at present of a third landing pad at LZ-1.

If the next mission is STP-2 NET Apr 30, will we see a third pad built for that one?

[old_sellsword]

Of course to state the obvious, there are no signs at present of a third landing pad at LZ-1.

If the next mission is STP-2 NET Apr 30, will we see a third pad built for that one?

They cancelled plans for a third pad and expanded the Dragon area instead. So three-core RTLS is no longer an option.

I really wish we could read the text in their graphic to better understand the elements. Has anyone seen an original for this floating around?

Larger version, plus a site plan attached.

[Lars-J]

Yep. They cold certainly add a 3rd landing pad in the future - but unlike hkultala, I see the FH demo configuration (boosters RTLS and core ASDS) as being the one that will fly the vast majority of FH's payloads.

[Lar]

I expect the proportion of all 3 RTLS to be far lower[1] than booster RTLS and center ASDS. Maybe instead of higher/lower it would be more accurate to speak of faster/slower velocity, as I do think that we'll see more lofted trajectories some times.

Since I expect pretty serious throttledown of the center as soon as a little mass is reduced I think I'm going to stick with my view that the boosters  will stage lower and slower than a single stick, most of the time.... If they had upsized S2, this would be pretty obviously true but even without I think it still is.

1 - possibly zero unless they build another landing pad or land two on one pad.

[hamerad]

I expect the proportion of all 3 RTLS to be far lower[1] than booster RTLS and center ASDS. Maybe instead of higher/lower it would be more accurate to speak of faster/slower velocity, as I do think that we'll see more lofted trajectories some times.

Since I expect pretty serious throttledown of the center as soon as a little mass is reduced I think I'm going to stick with my view that the boosters  will stage lower and slower than a single stick, most of the time.... If they had upsized S2, this would be pretty obviously true but even without I think it still is.

1 - possibly zero unless they build another landing pad or land two on one pad.

I would agree except for the option of having an ASDS just offshore. Just how close would the asds have to be to muddy the waters and make it an ASDS/RTLS

[Robotbeat]

If you watch some F9 launches like NROL-76, the booster and the upper stage coast along for several seconds with the booster broadside to the direction of travel, apparently under control of the cold gas thrusters. FH boosters will be higher, and the atmosphere density and drag halves about every 3 km at those altitudes.

I may be confused but I believe that since the FH is a 2.5 stage rocket the boosters will stage lower, not higher than an F9 S1. The center core will stage higher though, I think...

That assumption is correct for crossfeed, but not without it.
...

Unproven.

[Semmel]

It is pretty obvious that both options are technically possible. Faster and slower staging of the side boosters. Which ever occurs more often is like reading tea leaves and not worth the effort. The question you should be asking in the context of this thread is, does SpaceX want to be able to stage low or are they cool with always staging high? Because that has an impact on the separation method, which is the topic of this thread.

[Kaputnik]

Maybe they will want to look after thise centre cores, since they are specialised and thus rarer, and ASDS landing allows more prop for entry and landing burns, resulting in less damage to the stage.

[Pete]

The Falcon9 Heavy faces some very interesting challenges.
*it needs to liftoff without blowing the launch pad to high heaven. 27 Merlins in one place at one time is serious firepower.
*it has a less aerodynamically forgiving profile. It wants to have higher t/w ratio. The combination of these will make max-q very very interesting.
*did I mention how interesting the superimposed and impinging transonic/low supersonic shockwaves around three parallel cylinders will be?
*A tall narrow cylinder is naturally very strong only in its length. Having differential twr between the side boosters and the core imposes bending moments in exactly the direction that these cylinders are not strong.
*and lastly, the question that this thread is actually about: When you finally Do drop the side boosters, how do you ensure the separation is clean, does not induce transient loads that shatter everything, and ensures no re-contact between anything.

Fortunately, we have Elon Musk's statements on this matter to put our concerns to rest:
"I hope it makes it far enough away from the pad that it does not cause pad damage, I would consider even that a win"
and
"very exciting, major pucker factor, really"

[Jim]

1,  it needs to liftoff without blowing the launch pad to high heaven. 27 Merlins in one place at one time is serious firepower.
2.  it has a less aerodynamically forgiving profile. It wants to have higher t/w ratio. The combination of these will make max-q very very interesting.
3.  did I mention how interesting the superimposed and impinging transonic/low supersonic shockwaves around three parallel cylinders will be?
4.  A tall narrow cylinder is naturally very strong only in its length. Having differential twr between the side boosters and the core imposes bending moments in exactly the direction that these cylinders are not strong.

not really.  All this has been done many times before

1. Previous vehicles that used the pad had higher thrust. 

2 It can throttle and will, hence this will not be an issue. Core is going to throttle down soon after liftoff.   Max q will be similar to other vehicle.

3.  Not a big deal, see Titan III/IV, Delta IV Heavy, STS, etc

4.  There are none.  The attach points are below and above the stage/tanks.  The upper attach points put loads into each other and not the middle vehicle.