Author Topic: Reusability effect on costs  (Read 180083 times)

Offline RedLineTrain

  • Full Member
  • ****
  • Posts: 702
  • Liked: 413
  • Likes Given: 394
Re: Reusability effect on costs
« Reply #600 on: 07/11/2017 02:39 PM »
Those views don't seem to match up with SpaceX's stated annual manufacturing rate -- Cantrell suggests 10-12 while Shotwell states 20+.  Reusability with Blocks 3 and 4 would seem to allow a launch rate that is 2x or 3x 20+.

Offline envy887

  • Senior Member
  • *****
  • Posts: 4122
  • Liked: 2131
  • Likes Given: 1273
Re: Reusability effect on costs
« Reply #601 on: 07/11/2017 02:44 PM »
On Quora Jim Cantrell claims F9 booster re-use is primarily about increasing launch frequency to improve cash flow, as well as perfecting re-use technology for Mars:

https://www.quora.com/How-much-does-SpaceX-save-by-reusing-a-Falcon-rocket#

Claims very few boosters will re-used more than '3 or so' times, so seems to be ignoring (or doesn't believe?) SpaceX's claims about block 5 aim of 10 uses without refurbishment.

Here's the crux of his argument:

Quote from: Jim Cantrell
If you go through the R&D costs of developing a reusable launch vehicle, the opportunity costs (in terms of fuel used for return and the lost revenue opportunity for more payload to orbit) of returning the launcher first stage, and the costs of refurbishment between flights, generally accepted practice shows that you have to re-use the booster or launch vehicle 510 times before you make your money back if you account for all the costs.

So he's citing three reasons why they have to fly 5-10 times before saving any money:

1) R&D cost of developing reuse
2) Opportunity cost of the extra fuel and lower payload to orbit
3) Refurbishment cost between flights.

In the particular case of SpaceX, these arguments have been rebutted many times.

1) The R&D cost is a) already paid and b) can be recovered over the entire operational life of the rocket family. For a company that has already developed reusability, that's not an ongoing cost and doesn't factor into the cost differential between reusing and expending a rocket. The actual recovery of the cost depends on many future launches, so it's impossible to strictly ascribe a segment of the cost to a particular launch - it depends on the financial model used.

2) Pure $/kg to orbit comparisons are fundamentally flawed because payloads aren't fungible. The majority of payloads are sized so that SpaceX can launch them and recover the rocket, and so the payload reduction is completely irrelevant. The cost of fuel is negligible, and SpaceX can still launch large payloads expendable. (There is a similar argument that a smaller cheaper expendable rocket could launch the same payload to the same orbit, but this ignores the fact that such a rocket doesn't exist and actually does have a real payload reduction - it can't use recovery margins in an off-nominal mission or to lift a bigger payload).

3) Refurbishment is a real problem for many reusable systems, both real and proposed. It was the single biggest issue keeping the Shuttle from reaching it's initial goals. But SpaceX is working towards reducing refurbishment to being less than fuel costs, and even if they only get within 10x fuel costs there is potential for very significant savings, 50 to 70% of per-flight costs.

His arguments are somewhat applicable to investment-limited companies that have to show a profit in the near term while attempting to develop reuse. They are much less applicable to a privately company that has a solid financial standing and is willing to put both all its profits and some investment money into developing something with a long-term payoff.
« Last Edit: 07/11/2017 02:45 PM by envy887 »

Offline Ludus

  • Full Member
  • ****
  • Posts: 1057
  • Liked: 450
  • Likes Given: 227
Re: Reusability effect on costs
« Reply #602 on: 07/14/2017 03:25 AM »
ULA plans (hopes?) to develop a rocket with substantially less performance than New Glenn, based on the same booster engine, and they only aspire to recover the engines, whereas Blue will recover the whole booster stage. Smart? More like the art of the possible, where the possible is clearly not sufficient.

I actually don't understand what ULA is doing. Using an engine from BO plain does not make sense.

When BO starts reliably launching their own LV, why on Earth would they continue to sell their engine to a competitor? Out of their heart goodness? Surely not. At best (for ULA), they would continue to sell it at a premium, much above their internal cost. At "not best", they just sink ULA whenever they want by simply refusing to prolong the contract.

ULA needs the engine as a replacement for the RD-180 on Vulcan and in that role it can't be part of a fully reusable Booster. ULA doesn't even plan to try their "smart" reuse of just the engines until years after they start flying them expendable. So Blue Origin gets to make money much sooner than flying New Glenn, scale their engine production up more, and embed themselves in the conventional Space Industry with an Alabama factory and consequent political support. ULA will just throw away the engines with each use for years. When it does eventually attempt reuse it will be a partial and unproven method. Meanwhile Blue Origin will apply the same engines that are partially subsidized by ULA to its own fully reusable New Glenn, years before, while ULA is still throwing them away with no prospect of really competing on cost.

Online MikeAtkinson

  • Full Member
  • ****
  • Posts: 1754
  • Bracknell, England
  • Liked: 518
  • Likes Given: 66
Re: Reusability effect on costs
« Reply #603 on: 07/14/2017 05:06 AM »
(There is a similar argument that a smaller cheaper expendable rocket could launch the same payload to the same orbit, but this ignores the fact that such a rocket doesn't exist and actually does have a real payload reduction - it can't use recovery margins in an off-nominal mission or to lift a bigger payload).

A smaller cheap launcher does/did exist. F9 v1.1, in a apples to apples comparison SpaceX determined that going reusable reduced costs.

Offline guckyfan

  • Senior Member
  • *****
  • Posts: 6678
  • Germany
  • Liked: 1732
  • Likes Given: 1695
Re: Reusability effect on costs
« Reply #604 on: 07/14/2017 05:19 AM »
A smaller cheap launcher does/did exist. F9 v1.1, in a apples to apples comparison SpaceX determined that going reusable reduced costs.

How do you know the 1.1 was cheaper? I think it was said that production became more efficient with design changes.

Offline envy887

  • Senior Member
  • *****
  • Posts: 4122
  • Liked: 2131
  • Likes Given: 1273
Re: Reusability effect on costs
« Reply #605 on: 07/14/2017 01:00 PM »
(There is a similar argument that a smaller cheaper expendable rocket could launch the same payload to the same orbit, but this ignores the fact that such a rocket doesn't exist and actually does have a real payload reduction - it can't use recovery margins in an off-nominal mission or to lift a bigger payload).

A smaller cheap launcher does/did exist. F9 v1.1, in a apples to apples comparison SpaceX determined that going reusable reduced costs.

I doubt that v1.1 was cheaper per launch to build or operate. The booster is almost identically sized, and the rocket uses the same engines throughout. The second stage was slightly smaller, but I doubt that the tank stretch added a significant cost to v1.2. And subcooling probably didn't add much in terms of marginal hardware cost (it mostly added development cost).

v1.0 might have been lower marginal cost, as it was about half the size and was initially priced cheaper. But v1.0 payload to GTO was about 30% less the v1.2 with ASDS recovery and less than half of v1.2 Block 5 fully expendable. v1.0 fully expended was sub 4 tonnes and not in a sweet spot for GTO launches, while v1.2 with recovery is in at good performance level of 5.0 to 5.5 tonnes.

Online MikeAtkinson

  • Full Member
  • ****
  • Posts: 1754
  • Bracknell, England
  • Liked: 518
  • Likes Given: 66
Re: Reusability effect on costs
« Reply #606 on: 07/14/2017 01:34 PM »
A smaller cheap launcher does/did exist. F9 v1.1, in a apples to apples comparison SpaceX determined that going reusable reduced costs.

How do you know the 1.1 was cheaper? I think it was said that production became more efficient with design changes.

I never said it was cheaper, just cheap.

The point I was making is that the argument, that you could produce a smaller, cheaper non-reusable launcher, is directly contradicted by SpaceX. When they wanted to enhance F9 v1.1 they had three options,
1. basically leave it as is, just try and make it cheaper.
2. improve performance, but  leave as a non-reusable (thus saving a lot of money getting reusability working).
3. the path they have taken.

SpaceX with full access to cost, performance and reliability data decided that reusability was the lowest cost solution. So a direct apples to apples comparison shows that it is not possible to have a cheaper expendable rocket that can launch equivalent payloads, at least not in a design similar to F9.

Offline envy887

  • Senior Member
  • *****
  • Posts: 4122
  • Liked: 2131
  • Likes Given: 1273
Re: Reusability effect on costs
« Reply #607 on: 07/14/2017 02:21 PM »
A smaller cheap launcher does/did exist. F9 v1.1, in a apples to apples comparison SpaceX determined that going reusable reduced costs.

How do you know the 1.1 was cheaper? I think it was said that production became more efficient with design changes.

I never said it was cheaper, just cheap.

The point I was making is that the argument, that you could produce a smaller, cheaper non-reusable launcher, is directly contradicted by SpaceX. When they wanted to enhance F9 v1.1 they had three options,
1. basically leave it as is, just try and make it cheaper.
2. improve performance, but  leave as a non-reusable (thus saving a lot of money getting reusability working).
3. the path they have taken.

SpaceX with full access to cost, performance and reliability data decided that reusability was the lowest cost solution. So a direct apples to apples comparison shows that it is not possible to have a cheaper expendable rocket that can launch equivalent payloads, at least not in a design similar to F9.

v1.1 was already reusable, but I think your point applies just as well to v1.0 vs v1.1.

 The v1.0 was already small, and cheap enough as an expendable. SpaceX decided it would be cheaper per flight to make it bigger and reusable in v1.1, and they doubled down on that move with v1.2.

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 10123
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 6965
  • Likes Given: 4754
Re: Reusability effect on costs
« Reply #608 on: 07/14/2017 03:17 PM »
1.0 Didn't serve the market though. Not enough payload to GTO.
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Offline Lars-J

  • Senior Member
  • *****
  • Posts: 4037
  • California
  • Liked: 3331
  • Likes Given: 2097
Re: Reusability effect on costs
« Reply #609 on: 07/14/2017 04:11 PM »
The v1.0 was already small, and cheap enough as an expendable. SpaceX decided it would be cheaper per flight to make it bigger and reusable in v1.1, and they doubled down on that move with v1.2.

The v1.0 was very complex to build (they only ever built and launched two in a year - a total of 5 over three years), and had not so great performance. With the experience they now had, they went back to the drawing board and redesigned it from the ground up - v1.1.

This is one of the bravest and best decisions SpaceX ever did. The v1.1 was simpler and more reliable, easier to build, and had double the performance of the v1.0. (and even more now with v1.2/FT) Their launch cadence immediately jumped much higher. AND as a bonus... It also added re-usability features. (although not realized until v1.2)

« Last Edit: 07/14/2017 04:13 PM by Lars-J »

Offline sanman

  • Senior Member
  • *****
  • Posts: 4125
  • Liked: 574
  • Likes Given: 8
Re: Reusability effect on costs
« Reply #610 on: 08/23/2017 09:46 PM »
So I'd like to know how many times will the typical Falcon booster stage get re-used? How many flights in total can we expect from a typical Falcon booster? That's obviously going to be the biggest determinant on cost.


So far I've heard people here say that a booster would likely only be re-flown a couple of times max - so basically 3 flights total for a booster.
Musk and Bezos have thrown around the 747 analogy, saying that plane tickets would be a lot more expensive if you had to throw away the plane after each flight. Well, they'd still be pretty darn expensive if you have to throw away the plane after just 3 flights.

Are there any plans to go beyond just a couple of re-flights, and have something that can fly, say 10 times?

Offline Ictogan

  • Full Member
  • *
  • Posts: 128
  • Germany
  • Liked: 76
  • Likes Given: 107
Re: Reusability effect on costs
« Reply #611 on: 08/23/2017 09:55 PM »
So I'd like to know how many times will the typical Falcon booster stage get re-used? How many flights in total can we expect from a typical Falcon booster? That's obviously going to be the biggest determinant on cost.


So far I've heard people here say that a booster would likely only be re-flown a couple of times max - so basically 3 flights total for a booster.
Musk and Bezos have thrown around the 747 analogy, saying that plane tickets would be a lot more expensive if you had to throw away the plane after each flight. Well, they'd still be pretty darn expensive if you have to throw away the plane after just 3 flights.

Are there any plans to go beyond just a couple of re-flights, and have something that can fly, say 10 times?
I think someone from SX said that it would be max 3 flights for block 3 cores and about 10 flights(without major refurbishment) for block 5 cores.

Online oiorionsbelt

  • Full Member
  • ****
  • Posts: 1292
  • Liked: 534
  • Likes Given: 1226
Re: Reusability effect on costs
« Reply #612 on: 08/23/2017 11:05 PM »

I think someone from SX said that it would be max 3 flights for block 3 cores and about 10 flights(without major refurbishment) for block 5 cores.
It was Gwynne Shotwell on the Space Show and she said, it was more like inspection than refurbishment and then go fly and they hope to re fly the block 5's a dozen or so times.

Online oldAtlas_Eguy

  • Senior Member
  • *****
  • Posts: 3387
  • Florida
  • Liked: 1788
  • Likes Given: 215
Re: Reusability effect on costs
« Reply #613 on: 08/24/2017 04:27 PM »
After a while the biggest costs in refurbishment will not be labor but the building to store the boosters awaiting launch. 1 day of one person is ~$500 including overhead costs (management, support, benefits, etc.). Even at a crew of 10 for 1 week the total labor costs would be only $25,000. A crew of 20 for 2 weeks is just $100,000.

Added:
The estimate for floor-space costs is ~$5,000/month per booster for storage or work-space. A 6 months of storage could end being more than all total labor in "refurbishment".
« Last Edit: 08/24/2017 04:39 PM by oldAtlas_Eguy »

Offline sanman

  • Senior Member
  • *****
  • Posts: 4125
  • Liked: 574
  • Likes Given: 8
Re: Reusability effect on costs
« Reply #614 on: 08/26/2017 10:30 PM »
How do the costs go up when making the vehicle more and more reusable (ie. increasing the re-flight capability)?

Is there some magic number with an asymptote line, so that the costs skyrocket dramatically as you try to push the reusability (number of re-flights) beyond a certain point?

Offline FutureSpaceTourist

  • Senior Member
  • *****
  • Posts: 6217
  • UK
    • Plan 28
  • Liked: 5690
  • Likes Given: 1632
Re: Reusability effect on costs
« Reply #615 on: 08/26/2017 11:00 PM »
How do the costs go up when making the vehicle more and more reusable (ie. increasing the re-flight capability)?

Is there some magic number with an asymptote line, so that the costs skyrocket dramatically as you try to push the reusability (number of re-flights) beyond a certain point?

I imagine that would be very dependant on the rocket design. We know that Elon's original target for a 12 hour turnaround time between launches of the same booster would have required a fundamental redesign, whereas 24 hours didn't. I would expect a correlation between turnaround time and number of reflights.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 7414
  • Everyplaceelse
  • Liked: 1091
  • Likes Given: 7350
Re: Reusability effect on costs
« Reply #616 on: 08/28/2017 06:46 AM »
SpaceX with full access to cost, performance and reliability data decided that reusability was the lowest cost solution. So a direct apples to apples comparison shows that it is not possible to have a cheaper expendable rocket that can launch equivalent payloads, at least not in a design similar to F9.
That's one explanation of their behavior.

Another is that the market for that size of LV and the rate at which launches in that market were sold is too small for them to carry the expense of a 2nd production line (even a partial one, sharing as many parts as possible with F9).

Had someone come along and said "We'll buy your whole F1 mfg line, and all the R&D data" that would have put someone into the launch services business without the R&D costs and at a point in the market SX were no longer planning to compete in.  :(

IRL that didn't happen because AFAIK a)No such company contacted them and b) Even if they had so much of the F1 mfg hardware was re-assigned to F9 they would have stripped a fair bit of the factory. Nothing they couldn't have replaced, but at considerable cost and time.  :( 
BFS. The worlds first Methane fueled FFORSC engined CFRP structured A380 sized aerospaceplane tail sitter capable of flying in Earth and Mars atmospheres. BFR. The worlds biggest Methane fueled FFORSC engined CFRP structured booster for BFS. First flight to Mars by end of 2022. Forward looking statements. T&C apply. Believe no one. Run your own numbers. So, you are going to Mars to start a better life? Picture it in your mind. Now say what it is out loud.

Offline mvpel

  • Full Member
  • ****
  • Posts: 1124
  • New Hampshire
  • Liked: 1295
  • Likes Given: 1686
Re: Reusability effect on costs
« Reply #617 on: 08/31/2017 06:56 PM »
Their long-term aim is at least 100 reflights of each booster core, and I heard this directly from the SpaceX VP of business development at the CRS-8 launch viewing on the OMB-II verandah.
"Ugly programs are like ugly suspension bridges: they're much more liable to collapse than pretty ones, because the way humans (especially engineer-humans) perceive beauty is intimately related to our ability to process and understand complexity. A language that makes it hard to write elegant code makes it hard to write good code." - Eric S. Raymond

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 7414
  • Everyplaceelse
  • Liked: 1091
  • Likes Given: 7350
Re: Reusability effect on costs
« Reply #618 on: 08/31/2017 10:24 PM »
Their long-term aim is at least 100 reflights of each booster core, and I heard this directly from the SpaceX VP of business development at the CRS-8 launch viewing on the OMB-II verandah.
Well so far they Shotwell on the Space Show said they are looking to do 3 launches off a booster and V5 will be good for 10 without major work.

So either V5 will do that with a lot of refurb work above "inspection" or they are at least another generation to go from 10 to 100.  :(

TBH I think there will be at least 1, and possible 2 generations of F9 before they achieve that, possibly including another upgrade of the PICAX ablative, although apparently previous upgrades have focused on lowering the unit cost of the material. [EDIT rather than lowering erosion rates or density. The former is always good, the latter normally lowers heat conduction into the structure you're trying to protect. Unfortunately they tend to conflict, making ablative development tricky ]

Frankly I don't care what the reusability effects on costs are for SX. :(

I care on what the reusability effects on customers prices will be instead.
« Last Edit: 09/02/2017 08:34 AM by john smith 19 »
BFS. The worlds first Methane fueled FFORSC engined CFRP structured A380 sized aerospaceplane tail sitter capable of flying in Earth and Mars atmospheres. BFR. The worlds biggest Methane fueled FFORSC engined CFRP structured booster for BFS. First flight to Mars by end of 2022. Forward looking statements. T&C apply. Believe no one. Run your own numbers. So, you are going to Mars to start a better life? Picture it in your mind. Now say what it is out loud.

Offline JamesH65

  • Full Member
  • ****
  • Posts: 791
  • Liked: 511
  • Likes Given: 8
Re: Reusability effect on costs
« Reply #619 on: 09/01/2017 02:30 PM »
Their long-term aim is at least 100 reflights of each booster core, and I heard this directly from the SpaceX VP of business development at the CRS-8 launch viewing on the OMB-II verandah.
Well so far they Shotwell on the Space Show said they are looking to do 3 launches off a booster and V5 will be good for 10 without major

So either V5 will do that with a lot of refurb work above "inspection" or they are at least another generation to go from 10 to 100.  :(

I expect refurb after each ten, retirement after 100. No-one (except SpaceX) has ANY idea of the refurb costs after each ten flights. Forgive another car analogy, but I'd expect the refurb to be on the order of a new set of brake pads. Not an every day occurrence, but certainly not a transmission rebuild which is often not cost effective.

Tags: