Author Topic: When will reused first stages be the majority of launches?  (Read 12023 times)

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 5947
  • Germany
  • Liked: 1459
  • Likes Given: 1177
The serial full duration tests of the JCSAT-14 booster at McGregor  let me think about this. We are getting near to proof that the landed boosters are all capable of reflight. I move a reply in the SpaceX Manifest Updates and Discussion Thread 4 here because it is not appropriate there. I argue that contracts signed today for new boosters will not delay the transition. They can and will be renegotiated.

There is not going to be any sudden rush to buy reused cores, especially before one has flown.

.................

Let's give SpaceX a chance to actually qualify a booster for reflight and their customers a chance to get comfortable with the idea before we start assuming everything will suddenly start flying on reused cores.

Absolutely true, I agree. But I would bet (just a phrase, I don't bet) that in 2019 most launches will be on reused boosters including contracts already signed for new ones. The contracts will be renegotiated with reusable prices. By that time they will probably have enough cores in store that they don't need to build new ones before the Falcon family is phased out.

Offline meekGee

  • Senior Member
  • *****
  • Posts: 7257
  • N. California
  • Liked: 3639
  • Likes Given: 754
Here's a supporting observation.

With the projected flight rate - how come SpaceX hasn't built another in-ground test-stand?  With all the equipment already there, you could get a two-fer price...

As the numbers of pads increase, and flight rate, it seems logical they'll need to - unless they foresee a leveling-off of production rate.
ABCD - Always Be Counting Down

Offline Kansan52

  • Full Member
  • ****
  • Posts: 752
  • Liked: 218
  • Likes Given: 251
2019 seems correct. Look how long it was before the first successful landing and the number since then. The Hot Fire Test will prove the core for reuse, so skipping McGregor. Core production will be retasked for BFR.

But that means fewer fresh S1 cores to test. They may even skip McGregor and rely on the Hot Fire as the only test. Blows my idea that the Hot Fire Test will be going away.

McGregor will then build a new test stand for BFR.

Online AncientU

  • Senior Member
  • *****
  • Posts: 3872
  • Liked: 2335
  • Likes Given: 3196
I suspect that there have been a number of negotiations with SpaceX concerning using reflown cores at lower prices, including with many who are currently are manifested (on new cores).  We already know that SES was interested in being first, and maybe reusing the same core for a second flight (maybe they were just being optimistic/fan-boyish).  Iridium talked about using reflown cores after this contracted batch is lofted on new cores already purchased.  GS has mentioned multiple negotiations...

IMO, many orders will be announced in the days/weeks/months following the first successful reflight in September/October -- assuming of course that this first flight is successful.  Since the reflown cores have same capability as new cores, are designed and tested as reusable, have a reflight guarantee, and are $20M cheaper than the cheapest ride out there, there will be little to keep commercial flights on new cores a year into flights of reflown cores.  So, I'd agree that by 2019, a majority (>50%) of SpaceX commercial flights will be on reflown cores, and possibly a year earlier than that.  There is a distinct possibility that the majority of the World's commercial flights will be on reflown cores soon after that.
« Last Edit: 08/01/2016 08:49 PM by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline dorkmo

  • Full Member
  • ****
  • Posts: 609
  • Liked: 279
  • Likes Given: 727
i guess another behind the scenes question is "how much does the insurance cost for a reused booster?"

perhaps the full duration tests of the landed stage was done to reassure an insurance company? could that be true?

i wonder if spacex will have to insure their own launches if no one else offers a reasonable rate

Online AncientU

  • Senior Member
  • *****
  • Posts: 3872
  • Liked: 2335
  • Likes Given: 3196
i guess another behind the scenes question is "how much does the insurance cost for a reused booster?"

perhaps the full duration tests of the landed stage was done to reassure an insurance company? could that be true?

i wonder if spacex will have to insure their own launches if no one else offers a reasonable rate

It should follow the pattern seen with F9... after the first 5-10 flights, the insurance was not much more than that for Ariane -- today it is the same as Ariane, even with a failure last year and a new FT version with only 6(?) flights.  In this case, SpaceX is now a proven launch service, the F9 is a proven launch vehicle, and the second stage is unchanged.  Might be argued that going to the FT version was a bigger design iteration than reflying a core.

Shouldn't take the insurers long to run the numbers... they are not anchored to the way it has traditionally been done as are many of the manufacturers (who still don't believe it will fly successfully at all).
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 5947
  • Germany
  • Liked: 1459
  • Likes Given: 1177
i guess another behind the scenes question is "how much does the insurance cost for a reused booster?"

Insurance rates for Falcon 9 dropped surprisingly fast to near Ariane levels. I don't think it will take many successful flights for F9R to reach similar values. Big question is the first reflight.

Online AncientU

  • Senior Member
  • *****
  • Posts: 3872
  • Liked: 2335
  • Likes Given: 3196
<snip>
Absolutely true, I agree. But I would bet (just a phrase, I don't bet) that in 2019 most launches will be on reused boosters including contracts already signed for new ones. The contracts will be renegotiated with reusable prices. By that time they will probably have enough cores in store that they don't need to build new ones before the Falcon family is phased out.

While I agree with the majority of your OP, this final comment is worth a response.  IMO, for the foreseeable future (strange colloquialism since the future is not at all foreseeable), Falcon core production will probably continue at near its current rate of 20 or so per year to supply the stock of first-use-only customers such as the USG, thus also making up for losses from barge landings, etc.  I expect the Falcon family to fly hundreds of times, if not a thousand times, before it is phased out.  (I'm not in the camp of BFR replacing everything -- just as likely as semi-trucks replacing personal vehicles.)
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Online AncientU

  • Senior Member
  • *****
  • Posts: 3872
  • Liked: 2335
  • Likes Given: 3196
2019 seems correct. Look how long it was before the first successful landing and the number since then. The Hot Fire Test will prove the core for reuse, so skipping McGregor. Core production will be retasked for BFR.

But that means fewer fresh S1 cores to test. They may even skip McGregor and rely on the Hot Fire as the only test. Blows my idea that the Hot Fire Test will be going away.

McGregor will then build a new test stand for BFR.

BFR isn't going to be even remotely road transportable, so McGregor testing of a full core won't happen.
BFR engines (Raptors) are another story entirely...
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline JamesH65

  • Full Member
  • ****
  • Posts: 519
  • Liked: 280
  • Likes Given: 7
<snip>
Absolutely true, I agree. But I would bet (just a phrase, I don't bet) that in 2019 most launches will be on reused boosters including contracts already signed for new ones. The contracts will be renegotiated with reusable prices. By that time they will probably have enough cores in store that they don't need to build new ones before the Falcon family is phased out.

While I agree with the majority of your OP, this final comment is worth a response.  IMO, for the foreseeable future (strange colloquialism since the future is not at all foreseeable), Falcon core production will probably continue at near its current rate of 20 or so per year to supply the stock of first-use-only customers such as the USG, thus also making up for losses from barge landings, etc.  I expect the Falcon family to fly hundreds of times, if not a thousand times, before it is phased out.  (I'm not in the camp of BFR replacing everything -- just as likely as semi-trucks replacing personal vehicles.)

Are barge landings going to be any more 'lossy' than land landings? I suspect not. They have been testing this heavily, and more tests to go, so reliability should be pretty good. I don't think there is any inherent reasons why it cannot be close to as reliable as RTLS. Bearing in mind most FH slights will require centre core barge landings, its something they will really want to make very reliable indeed.

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 5947
  • Germany
  • Liked: 1459
  • Likes Given: 1177
Re: When will reused first stages be the majority of launches?
« Reply #10 on: 08/02/2016 10:17 AM »
Are barge landings going to be any more 'lossy' than land landings? I suspect not. They have been testing this heavily, and more tests to go, so reliability should be pretty good. I don't think there is any inherent reasons why it cannot be close to as reliable as RTLS. Bearing in mind most FH slights will require centre core barge landings, its something they will really want to make very reliable indeed.

It depends much on weather conditions. At the moment they need to fly their manifest as good and fast as they can. If in future they have a launch cadence that catches up with demand they can delay launches because of weather conditions at the barge landing site.

There were a few very bad weather situations during the first tries. Were they just bad luck or is this to be expected frequently?

Edit: They will do powered landings with crew and without parachute backup. They better be more than 99% reliable. Falcon 9 is their testbed to achieve that.
« Last Edit: 08/02/2016 10:20 AM by guckyfan »

Offline Alastor

  • Full Member
  • **
  • Posts: 205
  • Liked: 105
  • Likes Given: 230
Re: When will reused first stages be the majority of launches?
« Reply #11 on: 08/02/2016 10:32 AM »
Barge landing is inherently more difficult than land landings.
You are shooting at a moving (mainly on the vertical axis) and rolling target.

Therefore, you have a risk of tipping, even once you have successfully landed, as well as a risk of coming in slightly too high or too low. You are also in an environment where you can have to endure much higher winds (again, risk of tipping).
In other words, if the weather is too bad, you pretty much have to ditch the stage in the water. And if you have a high enough launch rate, you may HAVE to launch even though you know you cannot land.

Also, barge landings are used for more difficult missions (less margin) than land landing, since if you land on the barge, it is because you cannot hope to RTLS with the fuel you have. So more difficult conditions on the return path.

Pretty much everything seems more difficult with barge landing than land landing, so you can expect a lesser success rate.
Also, the data we have seems to agree with that analysis, with a 100% success rate on land landings so far (and much less so for barge landings).

Dragon landing probably will be more like land landing. You choose the spot, you choose the time, so you go for the bes conditions !

Offline raketa

  • Full Member
  • *
  • Posts: 171
  • Liked: 27
  • Likes Given: 15
Re: When will reused first stages be the majority of launches?
« Reply #12 on: 08/02/2016 11:24 AM »
First stage reuse will be 100% successful when falcon system will be replace with BFR+MCT. BFR will have enough power to land back on launch site. This system will be available before Mars adventure. Time frame for ale 2022-2026.

Offline JamesH65

  • Full Member
  • ****
  • Posts: 519
  • Liked: 280
  • Likes Given: 7
Re: When will reused first stages be the majority of launches?
« Reply #13 on: 08/02/2016 12:14 PM »
Barge landing is inherently more difficult than land landings.
You are shooting at a moving (mainly on the vertical axis) and rolling target.

Therefore, you have a risk of tipping, even once you have successfully landed, as well as a risk of coming in slightly too high or too low. You are also in an environment where you can have to endure much higher winds (again, risk of tipping).
In other words, if the weather is too bad, you pretty much have to ditch the stage in the water. And if you have a high enough launch rate, you may HAVE to launch even though you know you cannot land.

Also, barge landings are used for more difficult missions (less margin) than land landing, since if you land on the barge, it is because you cannot hope to RTLS with the fuel you have. So more difficult conditions on the return path.

Pretty much everything seems more difficult with barge landing than land landing, so you can expect a lesser success rate.
Also, the data we have seems to agree with that analysis, with a 100% success rate on land landings so far (and much less so for barge landings).

Dragon landing probably will be more like land landing. You choose the spot, you choose the time, so you go for the bes conditions !

Agreed barge landings are more difficult. I just think that they are not that much more difficult. After all, they have landed three already (albeit with varying levels of damage), and they are only at the start of their tests. Are we already at optimal barge landing? I seriously doubt it.

Am I right in thinking that FH centre cores will be landing on the barge with more reserves than F9 cores? That would give extra leeway, as would only launching when weather at LS was sane, which they should be able to do once their backlog is cleared and things are not so urgent.

Offline laszlo

  • Full Member
  • *
  • Posts: 118
  • Liked: 142
  • Likes Given: 17
Re: When will reused first stages be the majority of launches?
« Reply #14 on: 08/02/2016 12:18 PM »
At the risk of being jumped on and pounded into the ground as a "naysayer", engineering is about tested, measured and proven results. So far, while there are strong indications, everything involved with reusing an F9 is purely experimental. There is absolutely no evidence (yet) that it will both work and be economically viable. That won't happen until enough successful reflights are done to prove that it works and to be able to determine the actual financials. So far there hasn't been a single SpaceX reflight so no one knows what is actually possible vs. simply hoped for. Nor have there been enough landings to build up a true actuarial expectation of the rate of boosters returned in good enough shape to be re-used at economically viable refurbishment effort rates (landing with damage can make it cheaper to build a new booster rather than repair the old one). So this speculation is a bit premature, especially about the BFR which hasn't even cleared the tower, let alone returned to the launchpad.

<20 minutes of dodging brickbats>

SpaceX is doing exactly the right thing with all the tests. They are actually accumulating the evidence needed to make the business case. The full duration tests were done for the engineers, not the insurance companies. Even if all the testing shows no unmanageable problems, it could take til 2019 to know know if it makes economic sense to re-fly at all because of the need to accumulate a statistically significant actuarial data set. So having a majority of boosters re-used by then is optimistic.

Online AncientU

  • Senior Member
  • *****
  • Posts: 3872
  • Liked: 2335
  • Likes Given: 3196
Re: When will reused first stages be the majority of launches?
« Reply #15 on: 08/02/2016 12:36 PM »
At the risk of being jumped on and pounded into the ground as a "naysayer", engineering is about tested, measured and proven results. So far, while there are strong indications, everything involved with reusing an F9 is purely experimental. There is absolutely no evidence (yet) that it will both work and be economically viable. That won't happen until enough successful reflights are done to prove that it works and to be able to determine the actual financials. So far there hasn't been a single SpaceX reflight so no one knows what is actually possible vs. simply hoped for. Nor have there been enough landings to build up a true actuarial expectation of the rate of boosters returned in good enough shape to be re-used at economically viable refurbishment effort rates (landing with damage can make it cheaper to build a new booster rather than repair the old one). So this speculation is a bit premature, especially about the BFR which hasn't even cleared the tower, let alone returned to the launchpad.

<20 minutes of dodging brickbats>

SpaceX is doing exactly the right thing with all the tests. They are actually accumulating the evidence needed to make the business case. The full duration tests were done for the engineers, not the insurance companies. Even if all the testing shows no unmanageable problems, it could take til 2019 to know know if it makes economic sense to re-fly at all because of the need to accumulate a statistically significant actuarial data set. So having a majority of boosters re-used by then is optimistic.

The engineering for reusing cores happened a long time ago.  Now they have returned cores which are being tested to validate (prove) that engineering...  They are not at the beginning of the process, they are at the end.  And they are not engineering novices.

Similarly, they already have a business case and are selling re-flown cores at a 30% discount as we speak, with, I assume, appropriate disclaimers in case something completely unanticipated rears its ugly head.  Gwynne Shotwell isn't so naive as to head down this path without knowing it makes economic sense.

Since we are talking about the future, speculation is the coin of the realm, of course.  But speculating on whether good engineering has been done or if a closeable business case may someday emerge is assuming that someone hasn't done their homework.  (Betting against SpaceX is certainly the option taken by many.)  Yes, a statistically significant actuarial data set will take a while to accumulate... if you are the competition, best to not wait until that is in hand.
« Last Edit: 08/02/2016 12:57 PM by AncientU »
"If we shared everything [we are working on] people would think we are insane!"
-- SpaceX friend of mlindner

Offline ChrisWilson68

  • Senior Member
  • *****
  • Posts: 3334
  • Sunnyvale, CA
  • Liked: 1934
  • Likes Given: 2193
Re: When will reused first stages be the majority of launches?
« Reply #16 on: 08/02/2016 12:59 PM »
So far, while there are strong indications, everything involved with reusing an F9 is purely experimental. There is absolutely no evidence (yet) that it will both work and be economically viable. That won't happen until enough successful reflights are done to prove that it works and to be able to determine the actual financials.

You're misusing the phrase "absolutely no evidence".  You say "there are strong indications" in one sentence and "absolutely no evidence" in the next.

Evidence is anything that tends to indicate something.  It can be stronger or weaker.  You seem to be claiming there is "absolutely no evidence" until the very strongest possible evidence is completely assembled.

So, your use of the term "evidence" is inconsistent with the standard understanding of the term in the English language.

So much argument could be avoided if people would just use the standard meanings of words.

Offline Alastor

  • Full Member
  • **
  • Posts: 205
  • Liked: 105
  • Likes Given: 230
Re: When will reused first stages be the majority of launches?
« Reply #17 on: 08/02/2016 01:23 PM »
Barge landing is inherently more difficult than land landings.
Pretty much everything seems more difficult with barge landing than land landing, so you can expect a lesser success rate.
Also, the data we have seems to agree with that analysis, with a 100% success rate on land landings so far (and much less so for barge landings).

Agreed barge landings are more difficult. I just think that they are not that much more difficult. After all, they have landed three already (albeit with varying levels of damage), and they are only at the start of their tests. Are we already at optimal barge landing? I seriously doubt it.

Am I right in thinking that FH centre cores will be landing on the barge with more reserves than F9 cores? That would give extra leeway, as would only launching when weather at LS was sane, which they should be able to do once their backlog is cleared and things are not so urgent.

"Are we already at optimal barge landing?"
Of course not. There is a lot of improvement in front of us. I however argue that if we end up with a 99.99% success rate for RTLS, then we may have something like a 99.90% success rate for barge landings.
In other words : It is more difficult for inherent reasons and therefore will always be more risky and less successful. Even if it still may be very successful.

Now whether or not the weather at landing site may become a launch criterium, I think it may be a case of cost.
Examples (Disclamer : figures are completely bogus and just serve to demonstrate the kind of math that may be going on. Anyone quoting those figures as if they were actual ones is an idiot ! ) :
Let's say it costs 1 000 000 to ditch the stage in the ocean. Now it may cost a lot to the customer to delay the launch even if by one day. So let's say he evaluates his cost for delaying launch to 250 000 per day. Then he may agree to delay the launch by 4 days for bad weather at landing site, but if the bad weather at landing site lasts longer, he will consider that it has cost him as much as if he paid to go expandable, and ask you to launch anyways.
In the end, what matters when you launch a satellite is that you put it into orbit, and if you do so at a very low price, customers may agree to some delay for reasons that don't relate to successful deployment of the payload, because it is still cheaper, but if it becomes a burden, he may not be so happy about it.

Another reason to decide to go expandable would be if the next launch is becoming too close. It may be costly enough for SpaceX if they have to move to the left their whole manifest to justify launching expandable from time to time if delays start to accumulate.

All things considered, I think that even if the landing success rate in good conditions was 100% (which it obviously can never be), there still are good reasons for SpaceX to decide to sometimes go expandable. Especially if they continue to grow their manifest and have even less margin in their schedule than they currently have.

Offline the_other_Doug

  • Senior Member
  • *****
  • Posts: 2396
  • Minneapolis, MN
  • Liked: 1452
  • Likes Given: 2481
Re: When will reused first stages be the majority of launches?
« Reply #18 on: 08/02/2016 04:26 PM »
Let us not also forget that a vast majority of ASDS landings in the future will be for stages that have reached much higher speeds prior to MECO than the far more benign entry environments available to RTLS trajectories.

It's not really a matter of the ASDS landings being more difficult in terms of the targeting and landing operations.  They are riskier, if for any reason, because the stages must pass through a far hotter and less benign entry environment than RTLS'ed boosters will see, and it's the hot entry environment, moreso than the fuel margin, that has resulted in ASDS landing failures to date.

If there was a perfectly-placed island to do land landings on for GTO launches and FH cores, you would still, IMHO, see a higher failure rate in stage recovery downrange than you ever will on RTLS.  It ain't whether you land on a boat or on land, it's how much punishment the booster has to take to get there that makes the difference...
-Doug  (With my shield, not yet upon it)

Offline meekGee

  • Senior Member
  • *****
  • Posts: 7257
  • N. California
  • Liked: 3639
  • Likes Given: 754
Re: When will reused first stages be the majority of launches?
« Reply #19 on: 08/02/2016 04:35 PM »
... engineering is about tested, measured and proven results. So far, while there are strong indications, everything involved with reusing an F9 is purely experimental. There is absolutely no evidence (yet) that it will both work and be economically viable. That won't happen until enough successful reflights are done to prove that it works and to be able to determine the actual financials. So far there hasn't been a single SpaceX reflight so no one knows what is actually possible vs. simply hoped for. Nor have there been enough landings to build up a true actuarial expectation of the rate of boosters returned in good enough shape to be re-used at economically viable refurbishment effort rates (landing with damage can make it cheaper to build a new booster rather than repair the old one). So this speculation is a bit premature, especially about the BFR which hasn't even cleared the tower, let alone returned to the launchpad.


When engineering is done, there's never "tested proven results" of what you're engineering - only tested results of other things which you might rely on...

By the time F9R was engineered, they had plenty of tested proven results from test-stands and previous generation hardware, and they had analysis tools - and that's the best you can hope for.

There's a school of thought on this board that keeps emphasizing "it's not done yet".  We all know that.  We can all go hybernate and read the history books 10 years from now.

But this is a forward-looking discussion, about a forward-looking enterprise, and as such, so far, the trajectory of SpaceX (and dare I say, its competitors) has been following what many here predicted.  That's about as much as we're saying.  They performed well so far, and it's looking good moving forward.
ABCD - Always Be Counting Down

Tags: