That is interesting. I’m sorry I missed the recent Delta IV and Atlas V launches. I usually try to watch them, but apparently I wasn’t paying close enough attention lately.
(I tried to leave this comment at Space KSC, but none of the comment formats seemed to like me.)
It would be awesome if SpaceX gets the next AsiaSat off this month as well. Three launches in 40-some days would probably be way better than ever done before, no? And it gives me some hope that their launch manifest for this year might actually be possible. That’s optimistic, but it would be awesome.
Huh – what does he mean, it’s a “modern-era (2000-present)” record? What’s the real record?
Well, things were busy in the sixties.
I posted a similar comment on Jeff Foust’s Space Politics blog back in April. I noted, at that time, that SpaceX, having launched its CRS-3 Dragon-to-ISS mission on April 18, had set May 10 as the date for launching the first Orbcomm OG2 mission. I noted that this was just 22 days apart for consecutive missions launched from the same pad and that this was about half the least time ULA had ever managed and also nearly two weeks better than Arianespace had ever managed at Kourou.
The Orbcomm mission, regrettably, did not turn out to be the one to set this mark, but it’s satisfying to see that SpaceX managed to do it anyway, even if a bit later than planned. They’ve got two more upcoming missions – Asiasat 6 and CRS-4 – that are scheduled 20 and 18 days after their respective predecessor missions. After that, the Spaceflight Now schedule shows nothing until CRS-5 on Dec. 1. But I believe SpaceX intends to do the second Orbcomm OG2 mission and one other – according to Stephen Smith, a GEO comsat for Turkmenistan – between CRS-4 and CRS-5. That would maintain a pell-mell pace of launch operations by SpaceX and result in 9 missions for calendar 2014 by Dec. 1. If there is any other mission whose payload is in-hand by then, it would even be possible to get another bird off before year’s end and make it a 10-mission year. Personally, I would happily trade this notional last mission for the Dragon V2 pad abort test.
The last five months of 2014 could be unprecedentedly great ones for SpaceX.
Interesting, Vlad. I wasn’t previously aware of the Spaceflight Insider site.
Both schedules agree on SpaceX’s next two missions but diverge significantly after that. Neither shows anything by SpaceX between CRS-4 and CRS-5. If Stephen Smith is right, and I think he is, both schedules are missing two SpaceX missions between CRS-4 and CRS-5. Spaceflight Insider also has nothing by SpaceX between CRS-5 and CRS-6. Spaceflight Now shows a SpaceX mission for USAF between the CRS-5 & 6 flights. Again, both schedules are probably missing some SpaceX stuff in those gaps.
I was interested to note that both sites are in much better agreement about upcoming Atlas V missions. Spaceflight Insider has five such missions listed between now and next March. Spaceflight Now has the same five, plus a MUOS 3 mission its pegs for January departure, plus a seventh Atlas V mission pegged for June.
I’ve seen it said various places that the 15 RD-180’s ULA has in stock represent roughly a 2-year supply. If the “burn rate” of Atlas V missions is as portrayed on these two launch schedule web pages, then that two year figure would appear to be correct.
It will be interesting to see what, if any, changes occur to scheduled Atlas V missions – and how soon – in the event the pair of RD-180’s ULA is said to be expecting by the end of this month fail to arrive. The canaries in this particular coal mine would seem to be the GPS Block 2 missions. One of these, satellite no. 8 in the series, is on both schedules for October 29. Another mission, for satellite no. 10, is on Spaceflight Now for a June departure. Satellite no. 9, to be launched between these two, is slated by Spaceflight Now to head skyward on a Delta IV. If either or both of the currently scheduled Atlas V missions are delayed and/or replaced by Delta IV missions, that would be evidence that, formal announcement or no, ULA and USAF have gone into serious rationing mode for Atlas V’s.
Both sites surely have their own information gathering and validation protocols. Given the differences in the two schedules, their protocols are obviously not the same. It’s hard to account for the greater divergence of SpaceX-related entries between the two sites except by supposing SpaceX is simply harder to get hard information out of for missions beyond the next 30 – 60 days.
They are very pro NASA and the status quo with SLS/MPCV. I had tried posting advocating more commercial services and that NASA should utilize it more. Included the Space Act of 1958 that was ammened under President Reagan to include more commercial serives. Needless to say it didn’t play well there and none of my posts have seen the light of day. I just read articles there now and don’t bother to post.
Duly noted, Vlad. And thanks again for the info. As I already get more than my minimum daily requirement of legacy space cheerleading from commenters at NASA Watch, Space Politics, Behind the Black and a few other places, I’ll probably elect to follow your lead on staying away from Spaceflight Insider’s editorial content. That’s especially so if there is a local Ministry of Truth in the picture. But I’ll keep an eye on their mission schedule just to see how it differs from that of Spaceflight Now, which has been my go-to on such matters for quite awhile now.
Given SpaceX’s price point, even if they don’t achieve re-usability, they have the potential to dominate the launch market by virtue of being cheaper. A lot cheaper.
What’s standing in their way of doing so? Two things; their slow launch rate (which results in manifest slips and long waits for the customer) and the unproven nature of the Falcon 9 (thus, a risk factor).
Their performance this year, if we exclude the much-delayed Orbcomm launch, pretty much eliminates, for now, the slow launch rate problem. They seem to have successfully dealt with it, but it’ll take a couple more low-delay launches to make it firm.
That leaves the newness factor. That’s diminishing with every non-catastrophic launch, and IMHO, by the time they get to eight or nine F9 1.1 launches, it’ll be gone. (Yesterday’s launch was the 6th for f9 1.1, and the 11th for any F9 version.)
So, if they can keep up a roughly one a month flight rate, before the end of the year, they will, IMHO, have their competition over a barrel and soon be the dominant launch provider, and that’s assuming re-usability fails.
If I were Arienespace, ULA, Roscosmos, etc, I’d be very, very worried.
And all of the above is assuming SpaceX utterly fails on re-usability. But if they can make re-usability work (and it looks to me like they will) , it will increase their competitive advantage massively.
“but if they can make re-usability work (and it looks to me like they will)” I have no doubt that they will successfully soon land a first stage back on land. Does anyone have any clue, though, what kind of shape that first stage will be in? They haven’t yet managed to get one back even to look over (ignoring sea damage etc.) Is there reason to think that takeoff and re-entry isn’t very hard on a first stage? I’d like to be optimistic, but what is there to go on?
We know that the stage was fully functional right up to ocean landing.
Well, that certainly helps. I guess we’ll find out, though, whether the level of functionality enough to let the thing land, is good enough to take it right out and get it ready for refueling for the next flight. I think Shuttle needed to be taken apart and put back together. But we’ll see.
These AsiaSat flights use too much fuel to allow recovery of the stages. I guess they plan to use Falcon Heavy for them in the future and recover the stages, which would be cheaper, right?
If Falcon Heavy is basically three first stages plus one second stage, does that mean they recover everything except the payload? Or is there additional complex machinery to accomplish the crossfeeding which doesn’t get recovered?
Does anyone know: The center first stage, which is dropped last, and especially the second stage, are re-entering from much higher up. How much faster are they moving, and how much fuel penalty would that mean for each of them?
The Falcon Heavy can lift 21,200kg to GTO with cross-feed; the AsiaSat birds are under 5,000kg. That leaves a lot of performance margin. For that sort of mission they can dispense with cross-feed (which might complicate reuse, and which would leave the center core far downrange) and have plenty of fuel left over to return the three first stage cores to the launch site.
I don’t think SpaceX has done any testing of second stage reuse. They’ve said that the first stage is 70% of the cost of the Falcon 9, which suggests that first stage cores are >80% of the cost of a Falcon Heavy. First stage reuse is the low-hanging fruit.
If those costs are approximately right, and an expendable Falcon 9 costs X, a reusable Falcon Heavy would cost about .3X + 3*.1*.7X = ~.5X (assuming 10 uses of each first stage core). So about half the cost.
SpaceX hasn’t started on reuse of the upper stage yet. Falcon Heavy will allegedly be flown only in the reusability mode, though it’s hard to say if they’ll try to reuse every core stage, which might mean forgoing cross-feed. Even if that’s the case, the savings of reusing the cores is so massive that it makes sense to operate that way.
As for other complex bits, the rocket engines and propellant tanks dominate the cost of the vehicle. As far as I know, cross-feed doesn’t require anything super complicated, mostly just plumbing and interconnects, nothing like extra pumps.
It seems problematic to recover the second stage without a lot of ablative shielding to protect the stage during entry.
SpaceX intends to use a PICA-X heat shield on the top end of the 2nd stage to get it back from orbit intact. The concept videos showing 2nd-stage recovery have it landing, main engine down, on a pair of Super Draco-type thrusters rather than the Merlin 1-D vacuum engine. The recoverable 2nd stage would also be equipped with legs.
If the extra mass of the heat shield at the nose doesn’t at least roughly balance the mass of the engine + landing thrusters + landing legs at the rear, a re-entering 2nd stage might have a more than marginal tendency to want to swap ends on the way in. That would be major double-plus ungood. Making the heat shield so heavy would also be suboptimal from a payload penalty standpoint. So I’m guessing the reusable 2nd stage might need some extra attitude control hardware – maybe PICA-X versions of those paddle fins we saw recently on the Falcon 9R Dev 1 test article at McGregor?
Another possibility would be to gin up a much more muscular 2nd stage equipped with extended tankage and three Merlin 1-D’s instead of one. As one cannot accommodate three Merlin 1-D vacuum engine bells inside a circle the diameter of a Falcon 9, there would have to be three bulges at the base of such a stage to compensate. On re-entry, these would act as shuttlecock “feathers” to keep the stage from swapping ends. Build the bulges out of PICA-X and you’ve got something at least as reusable as a Dragon V2’s Super Draco sconces. If said bulges were split between the 2nd stage and the interstage, you’d also have useful “feathers” on the 1st-stage-plus-interstage unit. These might serve as stabilizer and anti-roll mechanisms on a par with either paddle fins or the current cold gas thrusters.
A 2nd-stage with three times the muscle would also be a growth path for Falcon 9 that would make it more than a match for any Atlas V or Delta IV configuration with strap-on solid boosters in both LEO and GTO lift capacity. It could also be used to raise Falcon Heavy’s performance specs into or beyond SLS Block 1 territory.
In the mid term SpaceX will switch from the Merlin engine to the Raptor engine. This means the first stage could have more performance than the Falcon 9 stage. I also suspect those engines will both be easier to reuse and have a lot better low-altitude performance because of the higher chamber pressure they can attain with staged combustion.
I don’t see the Raptor playing any role in Falcon 9’s future. It would be, physically, a much larger engine so it would hardly be plug-compatible with the Merlin 1-D. A single Raptor would be sea-level thrust equivalent to not quite seven of the Falcon 9’s current nine Merlin 1-D’s. That would both reduce payload and sacrifice Falcon 9’s enviable engine-out fault-tolerance. Adding a second Raptor probably can’t be done without both engine bells protruding beyond the current Falcon 9 diameter. Then there’s the fact of Raptor being a Metha-LOX engine instead of Kero-LOX. The Falcon 9R probably has some refinements still to come, but absent a more muscular upper stage, it’s pretty damned good as it stands.
The only role Raptor seems reasonably well-suited to, in a Falcon-series context, is as the power plant for a future, high-thrust, high-ISP upper stage for Falcon Heavy. If Elon sees profit to be had in building an SLS BLock 1 killer, this would appear to be the highest-performace path to achieving such at a modest price.
Mainly, though, the Raptor seems destined for a super-sized, Falcon-9-on-steroids reusable BFR core stage that will be flyable as either a singleton or as an ultra-BFR triple-core Heavy version with a 500+ tonne LEO lift capacity. SpaceX has no incentive to build an inferior Falcon 9 using Raptor. Raptor will, instead, power a heavy lifter that incorporates all F9R’s current advantages with an order of magnitude more lift capacity.
Dick, any reason not to put the landing legs and Super Dracos at the same end of the stage as the heatshield?
The landing legs could straighforwardly go on in either orientation. To put thrusters at the opposite end from the Merlin 1-D, though, would require either putting doors in the main heat shield – not generally a good idea – or going the Dragon V2 route and adding at least a pair of “sconces” to the 2nd stage so that the thrusters can fire at a slight angle on descent.
And, yes, the Dragon V2 incorporates tapered holes in its heat shield for the landing “feet.” But these feet are actual parts of the heat shield and are completely conformal with the rest of it during re-entry. They are also tapered slightly so that the hot gases pressing on them during re-entry just tighten their circumferential seal. That’s because they are attached to hydraulic/pneumatic landing struts that push these tapered plugs of heat shield material straight out on terminal approach to a landing. Putting rocket thrusters behind doors would either require a swing-in design, that would be much less intrinsically strong and gas-tight than Dragon V2’s landing leg design, or a swing-out design that would expose hinge hardware to re-entry plasma and radiant heat loads. Either would be a much dodgier design than the Dragon V2’s straighforward be-a-cork-in-a-bottle-during-re-entry-then-push-out-the-cork-from-inside-to-land approach.
So how many more sucessful launches until the AF’s reliability equations give a solution similar enough
to ULA’s to get them on board the bidding?
I think SpaceX is already well past that point. USAF’s certification requirements specify three successful launches, at least two of which have to be consecutive. I believe these criteria were established based on USAF experience with many previously developed families of satellite launchers as well as with the success/failure records of foreign launchers such as Soyuz, Proton and Ariane V. In most such prior cases, failures and partial failures have been concentrated toward the beginning of a rocket’s flight history. As Arizona CJ pointed out above, the Falcon 9v1.1 has now flown successfully six consecutive times and may well have done so 10 times or more by year’s end.
The Atlas V and Delta IV families have had one partial failure each, with respect to primary payloads, in their entire launch histories. In each case the primary payload failed to make it to the intended orbit, but did make it to some kind of orbit.
In the case of the Atlas V, the partial failure occurred on its 10th mission. If one counts the five Falcon 9 v1.0 missions in Falcon 9v1.1’s launch record, then the F9v1.1 has already surpassed Atlas V’s reliability record at a comparable point in its launch history. If one chooses not to so count the F9v1.0 missions, the F9v1.1 is mere months from notching 10 missions in its own right and its standing viv-a-vis the Atlas V’s early record will then be unambiguous for good or ill. But I don’t think making this distinction is defensible.
The Atlas V can fly in a variety of configurations having from zero up to five strap-on solid boosters. I think it would be fair to argue that Atlas V’s with one or more solids attached differ from unaugmented Atlas V’s to a greater degree than does the Falcon 9v1.1 from the Falcon 9v1.0. There have been 47 Atlas V missions to-date. 25 of these, just over half, have flown unaugmented. The one mission partial failure on the Atlas V’s track record occurred on one of these 25 unaugmented flights; the fifth such flight to be exact.
Adopting this standard for comparison, the Falcon 9v1.1, at six successful missions, has already bested the unaugmented Atlas V. So, for that matter, has the Falcon 9v1.0. If one takes the view that “Atlas V’s is Atlas V’s” then it is at least as reasonable to accept that “Falcon 9’s is Falcon 9’s” too. So we are back to 11 consecutive Falcon 9 successes from a standing start as opposed to the Atlas V’s partial failure on its tenth mission overall.
There are two similar distinctions that may, I think, be legitimately drawn in the case of Delta IV. The first is that between single-core missions unaugmented by strap-on solid boosters and those that used such boosters. In the case of Delta IV single-core missions, there have been a total of 20 over not quite 12 years, three without solids and 17 with. All have been successful.
If one was being pedantic, one could truthfully say that both Falcon 9v1.0 and Falcon 9v1.1, with five and six successful missions from their inceptions, respectively, have better demonstrated reliability records than the unaugmented Delta IV Medium because the latter has only flown three times.
Neither Falcon variant has yet notched 17 consecutive successes from a standing start, though, if one counts all Falcon 9’s, they are over half way along toward doing so. Even if the five Falcon 9v1.0’s are left off, the Falcon 9v1.1 is over a third of the way along toward doing so. Assuming no further mission failures for Falcon 9, the augmented Delta IV Medium’s record could be matched on or before the end of next year.
The only thing that might drive this contest, unresolved, into 2016 would be a sudden uptick in Delta IV Medium launch rate. If Atlas V’s suddenly become scarce owing to the highly likely cut-off of RD-180 engines by Russia, enough more Delta IV Mediums might well be launched before then end of 2015 to keep the Falcon 9, at least for a short while, at bay in the consecutive-successful-missions-from-inception derby.
As the one Delta IV to rack up a partial failure was also the first Delta IV Heavy ever launched, it is probably more reasonable to compare this vehicle’s track record to that of the yet-to-be-flown Falcon Heavy given that each is a three-core variant of a single-core vehicle with a thus-far perfect reliability record. Exceeding the early reliability record of the Delta IV Heavy will be accomplished if the first Falcon Heavy test flight simply manages to deliver whatever its payload turns out to be to the intended orbit. Again, that issue will almost certainly be settled well before the end of next year.
Both variants of the Falcon 9 already have much better early reliability records than Ariane 5, which has now racked up the longest string of consecutive mission successes among extant satellite launchers with the possible exception of Soyuz (haven’t had time to look into Soyuz in detail). Ariane 5’s first two missions were a failure and partial failure, respectively. These were followed by seven successes, another failure on mission 10, three more successes, another partial failure on mission 14, then the unbroken string of 60 mission successes that has continued down to the present day.
It is also fair to point out that, while widely criticized for its allegedly slow and trouble-prone launch tempo, SpaceX has still manged to get four missions up so far this year, has two more scheduled in the next five weeks and may get an additional two or three beyond those off before year’s end. That 60-consecutive-success Ariane 5 record has been achieved over a period of 12 years, an average of only five missions per year. The 20 Delta IV Medium, 7 Delta IV Heavy and 47 Atlas V missions to-date have also taken 12 years to accomplish. SpaceX has accomplished its 11 mission successes with Falcon 9 in a bit over four years.
From the article:
“Another record fell with this morning’s launch. The Cape pulled off three launches in eight days — the Delta IV on July 28, the Atlas V on August 2, and the Falcon 9 this morning. That’s three launches in eight days.”
Another one for the naysayers like Mark Whittington who claims it would impossible to launch fuel to a depot in a timely manner. He claims it would take a over a year.
Here is the chase plane video of the core water landing.
That is interesting. I’m sorry I missed the recent Delta IV and Atlas V launches. I usually try to watch them, but apparently I wasn’t paying close enough attention lately.
(I tried to leave this comment at Space KSC, but none of the comment formats seemed to like me.)
It would be awesome if SpaceX gets the next AsiaSat off this month as well. Three launches in 40-some days would probably be way better than ever done before, no? And it gives me some hope that their launch manifest for this year might actually be possible. That’s optimistic, but it would be awesome.
Huh – what does he mean, it’s a “modern-era (2000-present)” record? What’s the real record?
Well, things were busy in the sixties.
I posted a similar comment on Jeff Foust’s Space Politics blog back in April. I noted, at that time, that SpaceX, having launched its CRS-3 Dragon-to-ISS mission on April 18, had set May 10 as the date for launching the first Orbcomm OG2 mission. I noted that this was just 22 days apart for consecutive missions launched from the same pad and that this was about half the least time ULA had ever managed and also nearly two weeks better than Arianespace had ever managed at Kourou.
The Orbcomm mission, regrettably, did not turn out to be the one to set this mark, but it’s satisfying to see that SpaceX managed to do it anyway, even if a bit later than planned. They’ve got two more upcoming missions – Asiasat 6 and CRS-4 – that are scheduled 20 and 18 days after their respective predecessor missions. After that, the Spaceflight Now schedule shows nothing until CRS-5 on Dec. 1. But I believe SpaceX intends to do the second Orbcomm OG2 mission and one other – according to Stephen Smith, a GEO comsat for Turkmenistan – between CRS-4 and CRS-5. That would maintain a pell-mell pace of launch operations by SpaceX and result in 9 missions for calendar 2014 by Dec. 1. If there is any other mission whose payload is in-hand by then, it would even be possible to get another bird off before year’s end and make it a 10-mission year. Personally, I would happily trade this notional last mission for the Dragon V2 pad abort test.
The last five months of 2014 could be unprecedentedly great ones for SpaceX.
Spaceflight Insider has different times
http://www.spaceflightinsider.com/launch-schedule/
Interesting, Vlad. I wasn’t previously aware of the Spaceflight Insider site.
Both schedules agree on SpaceX’s next two missions but diverge significantly after that. Neither shows anything by SpaceX between CRS-4 and CRS-5. If Stephen Smith is right, and I think he is, both schedules are missing two SpaceX missions between CRS-4 and CRS-5. Spaceflight Insider also has nothing by SpaceX between CRS-5 and CRS-6. Spaceflight Now shows a SpaceX mission for USAF between the CRS-5 & 6 flights. Again, both schedules are probably missing some SpaceX stuff in those gaps.
I was interested to note that both sites are in much better agreement about upcoming Atlas V missions. Spaceflight Insider has five such missions listed between now and next March. Spaceflight Now has the same five, plus a MUOS 3 mission its pegs for January departure, plus a seventh Atlas V mission pegged for June.
I’ve seen it said various places that the 15 RD-180’s ULA has in stock represent roughly a 2-year supply. If the “burn rate” of Atlas V missions is as portrayed on these two launch schedule web pages, then that two year figure would appear to be correct.
It will be interesting to see what, if any, changes occur to scheduled Atlas V missions – and how soon – in the event the pair of RD-180’s ULA is said to be expecting by the end of this month fail to arrive. The canaries in this particular coal mine would seem to be the GPS Block 2 missions. One of these, satellite no. 8 in the series, is on both schedules for October 29. Another mission, for satellite no. 10, is on Spaceflight Now for a June departure. Satellite no. 9, to be launched between these two, is slated by Spaceflight Now to head skyward on a Delta IV. If either or both of the currently scheduled Atlas V missions are delayed and/or replaced by Delta IV missions, that would be evidence that, formal announcement or no, ULA and USAF have gone into serious rationing mode for Atlas V’s.
Both sites surely have their own information gathering and validation protocols. Given the differences in the two schedules, their protocols are obviously not the same. It’s hard to account for the greater divergence of SpaceX-related entries between the two sites except by supposing SpaceX is simply harder to get hard information out of for missions beyond the next 30 – 60 days.
They are very pro NASA and the status quo with SLS/MPCV. I had tried posting advocating more commercial services and that NASA should utilize it more. Included the Space Act of 1958 that was ammened under President Reagan to include more commercial serives. Needless to say it didn’t play well there and none of my posts have seen the light of day. I just read articles there now and don’t bother to post.
Duly noted, Vlad. And thanks again for the info. As I already get more than my minimum daily requirement of legacy space cheerleading from commenters at NASA Watch, Space Politics, Behind the Black and a few other places, I’ll probably elect to follow your lead on staying away from Spaceflight Insider’s editorial content. That’s especially so if there is a local Ministry of Truth in the picture. But I’ll keep an eye on their mission schedule just to see how it differs from that of Spaceflight Now, which has been my go-to on such matters for quite awhile now.
Given SpaceX’s price point, even if they don’t achieve re-usability, they have the potential to dominate the launch market by virtue of being cheaper. A lot cheaper.
What’s standing in their way of doing so? Two things; their slow launch rate (which results in manifest slips and long waits for the customer) and the unproven nature of the Falcon 9 (thus, a risk factor).
Their performance this year, if we exclude the much-delayed Orbcomm launch, pretty much eliminates, for now, the slow launch rate problem. They seem to have successfully dealt with it, but it’ll take a couple more low-delay launches to make it firm.
That leaves the newness factor. That’s diminishing with every non-catastrophic launch, and IMHO, by the time they get to eight or nine F9 1.1 launches, it’ll be gone. (Yesterday’s launch was the 6th for f9 1.1, and the 11th for any F9 version.)
So, if they can keep up a roughly one a month flight rate, before the end of the year, they will, IMHO, have their competition over a barrel and soon be the dominant launch provider, and that’s assuming re-usability fails.
If I were Arienespace, ULA, Roscosmos, etc, I’d be very, very worried.
And all of the above is assuming SpaceX utterly fails on re-usability. But if they can make re-usability work (and it looks to me like they will) , it will increase their competitive advantage massively.
“but if they can make re-usability work (and it looks to me like they will)” I have no doubt that they will successfully soon land a first stage back on land. Does anyone have any clue, though, what kind of shape that first stage will be in? They haven’t yet managed to get one back even to look over (ignoring sea damage etc.) Is there reason to think that takeoff and re-entry isn’t very hard on a first stage? I’d like to be optimistic, but what is there to go on?
We know that the stage was fully functional right up to ocean landing.
Well, that certainly helps. I guess we’ll find out, though, whether the level of functionality enough to let the thing land, is good enough to take it right out and get it ready for refueling for the next flight. I think Shuttle needed to be taken apart and put back together. But we’ll see.
These AsiaSat flights use too much fuel to allow recovery of the stages. I guess they plan to use Falcon Heavy for them in the future and recover the stages, which would be cheaper, right?
If Falcon Heavy is basically three first stages plus one second stage, does that mean they recover everything except the payload? Or is there additional complex machinery to accomplish the crossfeeding which doesn’t get recovered?
Does anyone know: The center first stage, which is dropped last, and especially the second stage, are re-entering from much higher up. How much faster are they moving, and how much fuel penalty would that mean for each of them?
The Falcon Heavy can lift 21,200kg to GTO with cross-feed; the AsiaSat birds are under 5,000kg. That leaves a lot of performance margin. For that sort of mission they can dispense with cross-feed (which might complicate reuse, and which would leave the center core far downrange) and have plenty of fuel left over to return the three first stage cores to the launch site.
I don’t think SpaceX has done any testing of second stage reuse. They’ve said that the first stage is 70% of the cost of the Falcon 9, which suggests that first stage cores are >80% of the cost of a Falcon Heavy. First stage reuse is the low-hanging fruit.
If those costs are approximately right, and an expendable Falcon 9 costs X, a reusable Falcon Heavy would cost about .3X + 3*.1*.7X = ~.5X (assuming 10 uses of each first stage core). So about half the cost.
SpaceX hasn’t started on reuse of the upper stage yet. Falcon Heavy will allegedly be flown only in the reusability mode, though it’s hard to say if they’ll try to reuse every core stage, which might mean forgoing cross-feed. Even if that’s the case, the savings of reusing the cores is so massive that it makes sense to operate that way.
As for other complex bits, the rocket engines and propellant tanks dominate the cost of the vehicle. As far as I know, cross-feed doesn’t require anything super complicated, mostly just plumbing and interconnects, nothing like extra pumps.
It seems problematic to recover the second stage without a lot of ablative shielding to protect the stage during entry.
SpaceX intends to use a PICA-X heat shield on the top end of the 2nd stage to get it back from orbit intact. The concept videos showing 2nd-stage recovery have it landing, main engine down, on a pair of Super Draco-type thrusters rather than the Merlin 1-D vacuum engine. The recoverable 2nd stage would also be equipped with legs.
If the extra mass of the heat shield at the nose doesn’t at least roughly balance the mass of the engine + landing thrusters + landing legs at the rear, a re-entering 2nd stage might have a more than marginal tendency to want to swap ends on the way in. That would be major double-plus ungood. Making the heat shield so heavy would also be suboptimal from a payload penalty standpoint. So I’m guessing the reusable 2nd stage might need some extra attitude control hardware – maybe PICA-X versions of those paddle fins we saw recently on the Falcon 9R Dev 1 test article at McGregor?
Another possibility would be to gin up a much more muscular 2nd stage equipped with extended tankage and three Merlin 1-D’s instead of one. As one cannot accommodate three Merlin 1-D vacuum engine bells inside a circle the diameter of a Falcon 9, there would have to be three bulges at the base of such a stage to compensate. On re-entry, these would act as shuttlecock “feathers” to keep the stage from swapping ends. Build the bulges out of PICA-X and you’ve got something at least as reusable as a Dragon V2’s Super Draco sconces. If said bulges were split between the 2nd stage and the interstage, you’d also have useful “feathers” on the 1st-stage-plus-interstage unit. These might serve as stabilizer and anti-roll mechanisms on a par with either paddle fins or the current cold gas thrusters.
A 2nd-stage with three times the muscle would also be a growth path for Falcon 9 that would make it more than a match for any Atlas V or Delta IV configuration with strap-on solid boosters in both LEO and GTO lift capacity. It could also be used to raise Falcon Heavy’s performance specs into or beyond SLS Block 1 territory.
In the mid term SpaceX will switch from the Merlin engine to the Raptor engine. This means the first stage could have more performance than the Falcon 9 stage. I also suspect those engines will both be easier to reuse and have a lot better low-altitude performance because of the higher chamber pressure they can attain with staged combustion.
I don’t see the Raptor playing any role in Falcon 9’s future. It would be, physically, a much larger engine so it would hardly be plug-compatible with the Merlin 1-D. A single Raptor would be sea-level thrust equivalent to not quite seven of the Falcon 9’s current nine Merlin 1-D’s. That would both reduce payload and sacrifice Falcon 9’s enviable engine-out fault-tolerance. Adding a second Raptor probably can’t be done without both engine bells protruding beyond the current Falcon 9 diameter. Then there’s the fact of Raptor being a Metha-LOX engine instead of Kero-LOX. The Falcon 9R probably has some refinements still to come, but absent a more muscular upper stage, it’s pretty damned good as it stands.
The only role Raptor seems reasonably well-suited to, in a Falcon-series context, is as the power plant for a future, high-thrust, high-ISP upper stage for Falcon Heavy. If Elon sees profit to be had in building an SLS BLock 1 killer, this would appear to be the highest-performace path to achieving such at a modest price.
Mainly, though, the Raptor seems destined for a super-sized, Falcon-9-on-steroids reusable BFR core stage that will be flyable as either a singleton or as an ultra-BFR triple-core Heavy version with a 500+ tonne LEO lift capacity. SpaceX has no incentive to build an inferior Falcon 9 using Raptor. Raptor will, instead, power a heavy lifter that incorporates all F9R’s current advantages with an order of magnitude more lift capacity.
I see that wikipedia
http://en.wikipedia.org/wiki/Falcon_Heavy#Reusable_technology_development
discusses some of my questions.
Dick, any reason not to put the landing legs and Super Dracos at the same end of the stage as the heatshield?
The landing legs could straighforwardly go on in either orientation. To put thrusters at the opposite end from the Merlin 1-D, though, would require either putting doors in the main heat shield – not generally a good idea – or going the Dragon V2 route and adding at least a pair of “sconces” to the 2nd stage so that the thrusters can fire at a slight angle on descent.
And, yes, the Dragon V2 incorporates tapered holes in its heat shield for the landing “feet.” But these feet are actual parts of the heat shield and are completely conformal with the rest of it during re-entry. They are also tapered slightly so that the hot gases pressing on them during re-entry just tighten their circumferential seal. That’s because they are attached to hydraulic/pneumatic landing struts that push these tapered plugs of heat shield material straight out on terminal approach to a landing. Putting rocket thrusters behind doors would either require a swing-in design, that would be much less intrinsically strong and gas-tight than Dragon V2’s landing leg design, or a swing-out design that would expose hinge hardware to re-entry plasma and radiant heat loads. Either would be a much dodgier design than the Dragon V2’s straighforward be-a-cork-in-a-bottle-during-re-entry-then-push-out-the-cork-from-inside-to-land approach.
So how many more sucessful launches until the AF’s reliability equations give a solution similar enough
to ULA’s to get them on board the bidding?
I think SpaceX is already well past that point. USAF’s certification requirements specify three successful launches, at least two of which have to be consecutive. I believe these criteria were established based on USAF experience with many previously developed families of satellite launchers as well as with the success/failure records of foreign launchers such as Soyuz, Proton and Ariane V. In most such prior cases, failures and partial failures have been concentrated toward the beginning of a rocket’s flight history. As Arizona CJ pointed out above, the Falcon 9v1.1 has now flown successfully six consecutive times and may well have done so 10 times or more by year’s end.
The Atlas V and Delta IV families have had one partial failure each, with respect to primary payloads, in their entire launch histories. In each case the primary payload failed to make it to the intended orbit, but did make it to some kind of orbit.
In the case of the Atlas V, the partial failure occurred on its 10th mission. If one counts the five Falcon 9 v1.0 missions in Falcon 9v1.1’s launch record, then the F9v1.1 has already surpassed Atlas V’s reliability record at a comparable point in its launch history. If one chooses not to so count the F9v1.0 missions, the F9v1.1 is mere months from notching 10 missions in its own right and its standing viv-a-vis the Atlas V’s early record will then be unambiguous for good or ill. But I don’t think making this distinction is defensible.
The Atlas V can fly in a variety of configurations having from zero up to five strap-on solid boosters. I think it would be fair to argue that Atlas V’s with one or more solids attached differ from unaugmented Atlas V’s to a greater degree than does the Falcon 9v1.1 from the Falcon 9v1.0. There have been 47 Atlas V missions to-date. 25 of these, just over half, have flown unaugmented. The one mission partial failure on the Atlas V’s track record occurred on one of these 25 unaugmented flights; the fifth such flight to be exact.
Adopting this standard for comparison, the Falcon 9v1.1, at six successful missions, has already bested the unaugmented Atlas V. So, for that matter, has the Falcon 9v1.0. If one takes the view that “Atlas V’s is Atlas V’s” then it is at least as reasonable to accept that “Falcon 9’s is Falcon 9’s” too. So we are back to 11 consecutive Falcon 9 successes from a standing start as opposed to the Atlas V’s partial failure on its tenth mission overall.
There are two similar distinctions that may, I think, be legitimately drawn in the case of Delta IV. The first is that between single-core missions unaugmented by strap-on solid boosters and those that used such boosters. In the case of Delta IV single-core missions, there have been a total of 20 over not quite 12 years, three without solids and 17 with. All have been successful.
If one was being pedantic, one could truthfully say that both Falcon 9v1.0 and Falcon 9v1.1, with five and six successful missions from their inceptions, respectively, have better demonstrated reliability records than the unaugmented Delta IV Medium because the latter has only flown three times.
Neither Falcon variant has yet notched 17 consecutive successes from a standing start, though, if one counts all Falcon 9’s, they are over half way along toward doing so. Even if the five Falcon 9v1.0’s are left off, the Falcon 9v1.1 is over a third of the way along toward doing so. Assuming no further mission failures for Falcon 9, the augmented Delta IV Medium’s record could be matched on or before the end of next year.
The only thing that might drive this contest, unresolved, into 2016 would be a sudden uptick in Delta IV Medium launch rate. If Atlas V’s suddenly become scarce owing to the highly likely cut-off of RD-180 engines by Russia, enough more Delta IV Mediums might well be launched before then end of 2015 to keep the Falcon 9, at least for a short while, at bay in the consecutive-successful-missions-from-inception derby.
As the one Delta IV to rack up a partial failure was also the first Delta IV Heavy ever launched, it is probably more reasonable to compare this vehicle’s track record to that of the yet-to-be-flown Falcon Heavy given that each is a three-core variant of a single-core vehicle with a thus-far perfect reliability record. Exceeding the early reliability record of the Delta IV Heavy will be accomplished if the first Falcon Heavy test flight simply manages to deliver whatever its payload turns out to be to the intended orbit. Again, that issue will almost certainly be settled well before the end of next year.
Both variants of the Falcon 9 already have much better early reliability records than Ariane 5, which has now racked up the longest string of consecutive mission successes among extant satellite launchers with the possible exception of Soyuz (haven’t had time to look into Soyuz in detail). Ariane 5’s first two missions were a failure and partial failure, respectively. These were followed by seven successes, another failure on mission 10, three more successes, another partial failure on mission 14, then the unbroken string of 60 mission successes that has continued down to the present day.
It is also fair to point out that, while widely criticized for its allegedly slow and trouble-prone launch tempo, SpaceX has still manged to get four missions up so far this year, has two more scheduled in the next five weeks and may get an additional two or three beyond those off before year’s end. That 60-consecutive-success Ariane 5 record has been achieved over a period of 12 years, an average of only five missions per year. The 20 Delta IV Medium, 7 Delta IV Heavy and 47 Atlas V missions to-date have also taken 12 years to accomplish. SpaceX has accomplished its 11 mission successes with Falcon 9 in a bit over four years.
From the article:
“Another record fell with this morning’s launch. The Cape pulled off three launches in eight days — the Delta IV on July 28, the Atlas V on August 2, and the Falcon 9 this morning. That’s three launches in eight days.”
Another one for the naysayers like Mark Whittington who claims it would impossible to launch fuel to a depot in a timely manner. He claims it would take a over a year.
Here is the chase plane video of the core water landing.
http://www.youtube.com/watch?feature=player_embedded&v=IBcj38ilNn4