Impact of SpaceX rideshare on small sat launchers market

[RedLineTrain]

Happy to hear your estimate of the marginal launch costs at a 100/yr launch rate.  Let's say $10 million?  $15 million?  It's got to be much less than two years ago at a 30/yr launch rate and no fairing reuse.

It should be fairly obvious that the floor price will be set by the price of a stg2, since that is the biggest non-reusable component.

When I did my little cost modelling game for this SX were saying that the proportion of the launch price of the US came to $17m.

Mass production (by rocket industry standards of "mass production") can lower that a bit, but how much? Perhaps Coastal_Ron could make a suggestion? The standard aircraft industry "learning curve" was a 15% fall in unit price per each doubling of production volume.

SpaceX have never said that an upper stage costs $17M. They have said it costs about $10M, with the fairings a further $6M if not reused.

I don't remember the $10 million upper stage reference.  Can you please point me to it?

[M.E.T.]

Happy to hear your estimate of the marginal launch costs at a 100/yr launch rate.  Let's say $10 million?  $15 million?  It's got to be much less than two years ago at a 30/yr launch rate and no fairing reuse.

It should be fairly obvious that the floor price will be set by the price of a stg2, since that is the biggest non-reusable component.

When I did my little cost modelling game for this SX were saying that the proportion of the launch price of the US came to $17m.

Mass production (by rocket industry standards of "mass production") can lower that a bit, but how much? Perhaps Coastal_Ron could make a suggestion? The standard aircraft industry "learning curve" was a 15% fall in unit price per each doubling of production volume.

SpaceX have never said that an upper stage costs $17M. They have said it costs about $10M, with the fairings a further $6M if not reused.

I don't remember the $10 million upper stage reference.  Can you please point me to it?

There’s been multiple references to this. Elon’s said it on Twitter. And in interviews. One might have been the Aviation Weekly interview with Irene Klotz a few years ago, which she linked on Twitter.

I mean, the $10M upper stage and $6M fairing costs are so well established that I thought it was common knowledge. But if I find the link to it I will update this post.

EDIT

Here’s the first link from a quick google. I’m sure there are plenty more:

https://www.elonx.net/how-much-does-it-cost-to-launch-a-reused-falcon-9-elon-musk-explains-why-reusability-is-worth-it/

[RedLineTrain]

Happy to hear your estimate of the marginal launch costs at a 100/yr launch rate.  Let's say $10 million?  $15 million?  It's got to be much less than two years ago at a 30/yr launch rate and no fairing reuse.

It should be fairly obvious that the floor price will be set by the price of a stg2, since that is the biggest non-reusable component.

When I did my little cost modelling game for this SX were saying that the proportion of the launch price of the US came to $17m.

Mass production (by rocket industry standards of "mass production") can lower that a bit, but how much? Perhaps Coastal_Ron could make a suggestion? The standard aircraft industry "learning curve" was a 15% fall in unit price per each doubling of production volume.

SpaceX have never said that an upper stage costs $17M. They have said it costs about $10M, with the fairings a further $6M if not reused.

I don't remember the $10 million upper stage reference.  Can you please point me to it?

There’s been multiple references to this. Elon’s said it on Twitter. And in interviews. One might have been the Aviation Weekly interview with Irene Klotz a few years ago, which she linked on Twitter.

I mean, the $10M upper stage and $6M fairing costs are so well established that I thought it was common knowledge. But if I find the link to it I will update this post.

OK, thank you.  Yes, this was the Irene Klotz podcast in May 2020.  In fact, I summarized it at the following post.

In Irene Klotz's interview of Musk in the run-up to DM-2, he puts the marginal cost of a Falcon 9 flight (1st stage and fairing reusable) at $15 million, of which $10 million is the upper stage.  Only $1 million for first stage refurbishment.

The discussion is about 4/10ths into the interview...

https://aviationweek.com/defense-space/space/podcast-interview-spacexs-elon-musk

Note that in 2020, the launch rate was one-fifth the current rate and one-sixth the rate that SpaceX plans for two months from now.  They were trying to catch the fairings in nets.

Reading between the lines of the recent history of Falcon 9, I am guessing that niobium was in short supply during covid, but is now more plentiful.  The use of the short nozzle on rideshare flights might be relevant here.

Relistening to the interview, I believe that he said that the booster requires "a quarter million" in refurbishment rather than $1 million in refurbishment.

Edit:  Putting it all together, I would be surprised to learn that the marginal cost for a rideshare flight on Falcon 9 as of the start of 2024 is as high as $10 million.  Ready fairing reuse, return to launch site, short nozzle, upper stage manufacturing rate increase, etc.

[M.E.T.]

Happy to hear your estimate of the marginal launch costs at a 100/yr launch rate.  Let's say $10 million?  $15 million?  It's got to be much less than two years ago at a 30/yr launch rate and no fairing reuse.

It should be fairly obvious that the floor price will be set by the price of a stg2, since that is the biggest non-reusable component.

When I did my little cost modelling game for this SX were saying that the proportion of the launch price of the US came to $17m.

Mass production (by rocket industry standards of "mass production") can lower that a bit, but how much? Perhaps Coastal_Ron could make a suggestion? The standard aircraft industry "learning curve" was a 15% fall in unit price per each doubling of production volume.

SpaceX have never said that an upper stage costs $17M. They have said it costs about $10M, with the fairings a further $6M if not reused.

I don't remember the $10 million upper stage reference.  Can you please point me to it?

There’s been multiple references to this. Elon’s said it on Twitter. And in interviews. One might have been the Aviation Weekly interview with Irene Klotz a few years ago, which she linked on Twitter.

I mean, the $10M upper stage and $6M fairing costs are so well established that I thought it was common knowledge. But if I find the link to it I will update this post.

OK, thank you.  Yes, this was the Irene Klotz podcast in May 2020.  In fact, I summarized it at the following post.

In Irene Klotz's interview of Musk in the run-up to DM-2, he puts the marginal cost of a Falcon 9 flight (1st stage and fairing reusable) at $15 million, of which $10 million is the upper stage.  Only $1 million for first stage refurbishment.

The discussion is about 4/10ths into the interview...

https://aviationweek.com/defense-space/space/podcast-interview-spacexs-elon-musk

Note that in 2020, the launch rate was one-fifth the current rate and one-sixth the rate that SpaceX plans for two months from now.  They were trying to catch the fairings in nets.

Reading between the lines of the recent history of Falcon 9, I am guessing that niobium was in short supply during covid, but is now more plentiful.  The use of the short nozzle on rideshare flights might be relevant here.

Relistening to the interview, I believe that he said that the booster requires "a quarter million" in refurbishment rather than $1 million in refurbishment.

Edit:  Putting it all together, I would be surprised to learn that the marginal cost for a rideshare flight on Falcon 9 as of the start of 2024 is as high as $10 million.  Ready fairing reuse, return to launch site, short nozzle, upper stage manufacturing rate increase, etc.

There has also been general inflation, though. Given the cost factors on both the up and downside I would probably be cautious and retain the $10M upper stage cost estimate as of today. Fairings probably still cost ~$6M to manufacture, so with 4 reuses that’s 1.5M per flight and throw in $500k fairing recovery and refurbishment cost to get to $2M fairing cost per flight.

So that’s the basis for my ~$12M estimate for upper stage and fairing.

Booster stage refurbishment probably increases with increased reuse as some engines eventually need to be swapped out (as Elon stated recently) so let’s make that a generous $1M per flight instead of the $250k he stated back then (including recovery operations).

Fuel is probably another ~$300k.

So that’s less than $13.5M, without launch day operations and booster amortisation. If the booster costs $35M, then 20 reuses gets us to less than $2M per flight. So now we’re at about $15M, excluding launch day operations. Make general operational launch costs ~$2M and we are at less than $18M TOTAL cost of launch for Falcon 9.

That’s assuming general overheads are already allocated into the cost of manufacture, refurbishment and launch.

That’s unbeatable.

[john smith 19]

So that’s less than $13.5M, without launch day operations and booster amortisation. If the booster costs $35M, then 20 reuses gets us to less than $2M per flight. So now we’re at about $15M, excluding launch day operations. Make general operational launch costs ~$2M and we are at less than $18M TOTAL cost of launch for Falcon 9.

Which puts their profit margin around what $42m? $45m? When a stage makes 20 launches.
Has any F9 stg1 reached that yet?

That’s unbeatable.

Profit margin so far yes. I doubt any ELV is walking away with a 233% profit margin.                     

Price? Not so much. Which is why SX has (if it gets to 20 launches) a 233% profit margin.

We might be finally at the point where people will start looking seriously at reusability, either partial or full.

Currently I count 3, possibly 4 routes for this.

[TrevorMonty]

So that’s less than $13.5M, without launch day operations and booster amortisation. If the booster costs $35M, then 20 reuses gets us to less than $2M per flight. So now we’re at about $15M, excluding launch day operations. Make general operational launch costs ~$2M and we are at less than $18M TOTAL cost of launch for Falcon 9.

Which puts their profit margin around what $42m? $45m? When a stage makes 20 launches.
Has any F9 stg1 reached that yet?

That’s unbeatable.

Profit margin so far yes. I doubt any ELV is walking away with a 233% profit margin.                     

Price? Not so much. Which is why SX has (if it gets to 20 launches) a 233% profit margin.

We might be finally at the point where people will start looking seriously at reusability, either partial or full.

Currently I count 3, possibly 4 routes for this.

Most new LVs in development have some reuseability, some will be expendable initially. Firefly MLV, RL Neutron, Relativity Terran R, Stoke Nova, Blue New Glenn even ULA Vulcan. 

[Zed_Noir]

<snip>
We might be finally at the point where people will start looking seriously at reusability, either partial or full.

Currently I count 3, possibly 4 routes for this.

Most new LVs in development have some reuseability, some will be expendable initially. Firefly MLV, RL Neutron, Relativity Terran R, Stoke Nova, Blue New Glenn even ULA Vulcan.

Other entries to the reusable launch provider market have to be quick in entering the market. Before the market leader gobbles up the market with an even cheaper reusable launcher that have greater payload capacity.

[TrevorMonty]

&lt;snip&gt;
We might be finally at the point where people will start looking seriously at reusability, either partial or full.

Currently I count 3, possibly 4 routes for this.

Most new LVs in development have some reuseability, some will be expendable initially. Firefly MLV, RL Neutron, Relativity Terran R, Stoke Nova, Blue New Glenn even ULA Vulcan.

Other entries to the reusable launch provider market have to be quick in entering the market. Before the market leader gobbles up the market with an even cheaper reusable launcher that have greater payload capacity.

SS will cheaper per kg but not per launch, unless SpaceX wants to run it at a loss.

[DanClemmensen]

SS will cheaper per kg but not per launch, unless SpaceX wants to run it at a loss.

You may be correct, but it's not certain. SpaceX appears to be going to great lengths to reduce the operations cost per launch of the SH/SS, with RTLS and little or no refurbishment and highly automated processing. I don't think we will know the absolute cost per launch until each of the contenders achieves full operational status. The low per-unit manufacturing costs also help, but not as much as operating efficiencies, especially with very high reuse.

Eventually the costs of methane and LOX will become important and may even dominate, but they will stabilize and then begin to come down as the energy market stabilizes.

[john smith 19]

Most new LVs in development have some reuseability, some will be expendable initially. Firefly MLV, RL Neutron, Relativity Terran R, Stoke Nova, Blue New Glenn even ULA Vulcan.

No. They have a claimed capability for (some) reusability, at some point in their evolution. As in fact does SS.

Right now only SX with F9 stg 1 are actually doing this. RL are doing some limited reuse, which should be be lowering their costs and I hope they move forward with full reuse of Electron stg1.

 Stoke and Blue look committed to full reuse, but Blue is looking more and more like a high end tax writeoff scheme to me. They've spent an awful lot of time and money and not really delivered that much

The rest sounds like PR to me. Vulcan starts from the premise of the smallest core hardware, then add SRB's. I doubt they've even started installing the hooks for SMART (IE doing the FEA and CAD for where to put the fittings when the do build an engine pod so they are ready).

[TrevorMonty]

Most new LVs in development have some reuseability, some will be expendable initially. Firefly MLV, RL Neutron, Relativity Terran R, Stoke Nova, Blue New Glenn even ULA Vulcan.

No. They have a claimed capability for (some) reusability, at some point in their evolution. As in fact does SS.

Right now only SX with F9 stg 1 are actually doing this. RL are doing some limited reuse, which should be be lowering their costs and I hope they move forward with full reuse of Electron stg1.

 Stoke and Blue look committed to full reuse, but Blue is looking more and more like a high end tax writeoff scheme to me. They've spent an awful lot of time and money and not really delivered that much

The rest sounds like PR to me. Vulcan starts from the premise of the smallest core hardware, then add SRB's. I doubt they've even started installing the hooks for SMART (IE doing the FEA and CAD for where to put the fittings when the do build an engine pod so they are ready).

Your reply is out if content. Copy all of post 326 then see if it makes sense.

[john smith 19]

Other entries to the reusable launch provider market have to be quick in entering the market. Before the market leader gobbles up the market with an even cheaper reusable launcher that have greater payload capacity.

Jess Sponable , who headed up the DCX project, said of development projects that you should ignore the competition claims of what they are going to do and focus on getting your project out the door at your specifications.

Then we see how your real vehicle compares against the competitors (possibly still mythical) vehicle.

WRT to thread title I'm sure the rideshare deal has and will kill off a few competitors. In some ways the game has not changed.

There's that old line about "We did not come to play. We came to win."

It also shows what will  make SX lower it's prices. Actual competition. Without it you launch on their schedule, at their prices.

[Zed_Noir]

<snip>
It also shows what will  make SX lower it's prices. Actual competition. Without it you launch on their schedule, at their prices.

Every drop in price by SpaceX will be that much harder for the competition to enter the market or get a meaningful return of investment in a timely manner.

Yes, I am advocating the Highlander movie meme of "There can be Only One" to SpaceX and it's so called competitors. 

[john smith 19]

Every drop in price by SpaceX will be that much harder for the competition to enter the market or get a meaningful return of investment in a timely manner.

Not necessarily.

The real payoff for initial investors is the IPO. the question then becomes is there a path that looks plausible to making a profit on the business so there are actual dividends, or a stock price rise, at the end so new investors will buy the stock?

Yes, I am advocating the Highlander movie meme of "There can be Only One" to SpaceX and it's so called competitors. 

Unlikely to happen as the USG will  always want at least 2 NSS suppliers.  It could then be argued that SX is not a monopoly, in the same way that Microsoft does not have a monopoly of desktop OS supply.

Not having an actual monopoly does not mean having an effective monopoly over a pretty large sector of the market.

What it will do is force a step change in what a new LV mfg/operator is proposing to investors in order to compete. Currently I can think of 3, possibly 4 architectures that offer such a possibility.

What I think SX's efforts to make F9 stg 2 reusability work have shown that trying to retrofit full reusability to a conventional VTO TSTO rocket doesn't work.

Stg 2 reusability has to be designed in from day one. Hoping "Something will turn up" is not a good long term plan. 

[DanClemmensen]

Yes, I am advocating the Highlander movie meme of "There can be Only One" to SpaceX and it's so called competitors. 

Unlikely to happen as the USG will  always want at least 2 NSS suppliers.  It could then be argued that SX is not a monopoly, in the same way that Microsoft does not have a monopoly of desktop OS supply.

What NSS "wants" may differ from what they will settle for. What they truly want is what their mission requires: assured access to space. If all else fails they will use a single supplier and then do what they can to see to it that their supplier survives.  This is what they did to force the shotgun marriage that created ULA, who became the single supplier from about 2006 until about 2018.  Sure, opening competition is (I think) more robust, and nurturing small launcher companies is a worthy long-term insurance strategy, but assured access to space is fundamental to the mission.

[john smith 19]

What NSS "wants" may differ from what they will settle for. What they truly want is what their mission requires: assured access to space. If all else fails they will use a single supplier and then do what they can to see to it that their supplier survives.  This is what they did to force the shotgun marriage that created ULA, who became the single supplier from about 2006 until about 2018.  Sure, opening competition is (I think) more robust, and nurturing small launcher companies is a worthy long-term insurance strategy, but assured access to space is fundamental to the mission.

Good point. And I'd bet they'll keep on bailing out ULA to do so. 

But in fact no current or previous supplier has ever given them the sort of "on-demand" access that they truly want.

For that they'd have to be able to buy a fully RLV at their site(s), launching on their schedule.

[TrevorMonty]

What NSS "wants" may differ from what they will settle for. What they truly want is what their mission requires: assured access to space. If all else fails they will use a single supplier and then do what they can to see to it that their supplier survives.  This is what they did to force the shotgun marriage that created ULA, who became the single supplier from about 2006 until about 2018.  Sure, opening competition is (I think) more robust, and nurturing small launcher companies is a worthy long-term insurance strategy, but assured access to space is fundamental to the mission.

Good point. And I'd bet they'll keep on bailing out ULA to do so. 

But in fact no current or previous supplier has ever given them the sort of "on-demand" access that they truly want.

For that they'd have to be able to buy a fully RLV at their site(s), launching on their schedule.

On demand launch means they have satellite in storage ready to go. With NSSL the satellites a more often worth far more than LV. 
On demand is really only useful for smaller constellation satellites and only likely to be launching one or two at time to fill hole.

[DanClemmensen]

What NSS "wants" may differ from what they will settle for. What they truly want is what their mission requires: assured access to space. If all else fails they will use a single supplier and then do what they can to see to it that their supplier survives.  This is what they did to force the shotgun marriage that created ULA, who became the single supplier from about 2006 until about 2018.  Sure, opening competition is (I think) more robust, and nurturing small launcher companies is a worthy long-term insurance strategy, but assured access to space is fundamental to the mission.

Good point. And I'd bet they'll keep on bailing out ULA to do so. 

But in fact no current or previous supplier has ever given them the sort of "on-demand" access that they truly want.

For that they'd have to be able to buy a fully RLV at their site(s), launching on their schedule.

On demand launch means they have satellite in storage ready to go. With NSSL the satellites a more often worth far more than LV. 
On demand is really only useful for smaller constellation satellites and only likely to be launching one or two at time to fill hole.

I speculate that SpaceX could offer an on-demand service if asked. They would basically need to keep one ready-to-go F9 in the queue, replacing it every time they do another Starlink launch. If/when NSSL demands a mission, bump the Starlink launch and fly the NSSL mission. Logistics would be a PITA so SpaceX would charge a monthly fee for the service whether or not a launch occurred. They might also need additional floor space at Roberts Road or Hangar X or both, to allow for queueing. Again, they would charge for it. NSSL would also need to arrange for preemptive use of the range on short notice if the next preempt-able F9 launch is too far in the future.

[john smith 19]

On demand launch means they have satellite in storage ready to go. With NSSL the satellites a more often worth far more than LV. 
On demand is really only useful for smaller constellation satellites and only likely to be launching one or two at time to fill hole.

Depends on what part of the NSS mission you're talking about. 

In some cases it's one off satellites, in others it's replacing a unit of a failed constellation, or adding to that to enhance coverage for surveillance or comms. 

[john smith 19]

I speculate that SpaceX could offer an on-demand service if asked. They would basically need to keep one ready-to-go F9 in the queue, replacing it every time they do another Starlink launch. If/when NSSL demands a mission, bump the Starlink launch and fly the NSSL mission. Logistics would be a PITA so SpaceX would charge a monthly fee for the service whether or not a launch occurred. They might also need additional floor space at Roberts Road or Hangar X or both, to allow for queueing. Again, they would charge for it. NSSL would also need to arrange for preemptive use of the range on short notice if the next preempt-able F9 launch is too far in the future.

It's not quite that simple.

Every payload is subject to a Coupled Loads Analysis, to ensure the payload isn't shaken to bits on the ascent and it can't excite the rocket into a vibration that shakes it to bits.

If SX wasn't the original launch provider that's going to be time consuming. It typically takes 3 iterations of this process to refine a payload design that will survive the launch and won't harm the rocket. This process is AFAIK unique to rocket launch. I'm unaware of large payloads on aircraft going through anything as detailed as CLA but I'm not sure if that's because no aircraft operates through 23 Mach numbers or if no aircraft has the levels of vibration or noise a rocket imposes on the payload.