Reusability effect on costs

[Radical_Ignorant]

I've seen discussion about this spread around multiple topics. Please kill this topic if there is already one but I couldn't find it on my phone browser.

So there are various guesses but often vehicle price is guessed as 70% first stage, 30% upper stage. So initial impression is that flying reused booster can reduce launch cost by 70%. Of course that's not true, because there is refrubrishment cost, launch operation cost,  payload integration cost etc. Still it sounds like if reusing 1st stage could reduce cost by about 50%. And we still are not talking about price as it makes no sense for Space X to reduce their profit from service provided so it is about cost reduction which is different from price reduction.
Let's assume that this 50% cost price is good enough to include price of stage if split among multiple launches so it's 50% cost reduction for all launches.

Is that the case? No. Because building stage is only part of it's cost. There is also R&D cost per stage which is independent from if stage is build or not. There are for sure people here who know how  big part is this. But please keep in mind it's not development of Falcon9 cost. It's quite pernament cost of keeping engineering department in house. Which is used to improve technology continuesly. And I believe it's quite a cost. So Space X needs to earn enough by their services to pay for it. Since they are continuesly innovating this can be treated as kind of fixed cost indendent from if rockets are build or not. This cost could be reduced once tech is developed but nobody (except some imaginary investors and short sighted clients) really want Space X to be scalled down.

Finally there is keeping production line cost. Those are employees which are not hired by temporary work agency. Nobody want SpaceX to lose higly qualified work force.

So finally my reasoning is that reusing (in short term) saves very little. Some raw materials and outsorced parts, but those are not many.

What reusability allows is to use resurces currently engaged in building stages to be used for other purposes. If there is market increase  some of those can be transferred to build more second stages. Now for the same operational cost there can be more launches and there ara real savings.

But if launch rate doesn't grow enough or SpaceX doesn't  scale down there is only limited saving.

So assuming reusability is given and SpaceX won't scale down, which I believe are true assumtions, cost reduction allowed by this humongous achievement is dependent from demand side and can be very small if there is no increase in launch rate.

Space X has some backlog so some increase is possible. But how big is that?

And am I missing something? Maybe raw materials and outsorced parts are really costly? Or my reasoning is flawed?

[bstrong]

There have been lots of discussions on this, and pretty much everyone agrees that you need a high launch rate to justify reuse. The main disagreement is over what exactly that launch rate is and how likely SpaceX is to achieve it.

I recommend these threads:

http://forum.nasaspaceflight.com/index.php?topic=35829.0
http://forum.nasaspaceflight.com/index.php?topic=37390.0

[Radical_Ignorant]

Sorry - my fault, or mobile opera to have something to blame :p
Economics of reusability is exactly the thread which I failed to found. Thanks a lot for pointing me there.

And this thread probably should be removed since it's duplicate to keep things clear. I couldn't do that myself.

[Lar]

The Economics of Reusability topic is old, last posted in 2014. It's a good topic to review for context, but probably since it's locked, discussion should continue here[1].

1 - Ignore the fact that I merged them together, realised it was old and locked, and then unmerged them. Just ignore it, I say!!!

[AncientU]

Gwynne Shotwell stated that a reused core would have a price reduced by 30%.  Since she seemed to be talking about now, I'd assume that she was talking about the present launch rate and manifest.  This should be a good price point to start the discussion.

Assuming the reused vehicle costs zero to refurbish*, the rest of launch costs plus profit (without considering the first stage) equal that 70% remaining -- nominally $43M.  If refurbishment costs are x, then the rest of launch costs are $43M minus x minus profit.

Note: Some say that SpaceX isn't making a profit, and others say they are losing money on each launch, so profit can be thought of as positive, zero, or negative.

* Recovery costs are part of 'refurbish'

[SoulWager]

The key issue has always been elasticity of demand, and the key issues for elasticity of demand are risk tolerance and time. Even a 50% discount is still be a lot of money for a startup.

I think SpaceX may need to fill its manifest gaps with home grown payloads while waiting for the satellite manufacturers to catch up. Hopefully that satellite internet thing works out.

[guckyfan]

There have been lots of discussions on this, and pretty much everyone agrees that you need a high launch rate to justify reuse.

I have to disagree on this. It's an old assumption, based on the expectation that a reusable design is a lot more expensive than an expendable one, requiring a lot of flights to recover cost. The Falcon reusable family is barely more expensive than expendable. It makes money with the first reuse.

[abaddon]

It makes money with the first reuse.

Assuming cost to recover & refurbish is lower than cost to manufacture, yes.  I have a really hard time seeing them burn $25-$30 million or whatever to recover & refurbish a first stage, so I agree.

[Rocket Science]

I'm just going to add a thought. Just because a stage it deemed "reusable" and launched it may not be "reliable"... One needs to factor this into the cost structure estimates...

[envy887]

I'm just going to add a thought. Just because a stage it deemed "reusable" and launched it may not be "reliable"... One needs to factor this into the cost structure estimates...

If rocket failures form a bathtub curve the reused stages will be more reliable than new ones, at least for a few launches.

[Dudely]

It makes money with the first reuse.

Assuming cost to recover & refurbish is lower than cost to manufacture, yes.  I have a really hard time seeing them burn $25-$30 million or whatever to recover & refurbish a first stage, so I agree.

A stage costs no more than $15 million. We know this because expendable FH costs 30 million > expendable F9.

If recovery and refurbishment cost 3 million they have saved roughly 80% of their costs. They could then lower the price to about 50 million and have the same profit per mission.

I have heard Gwyne say they expect to go as low as 43 million for reused flights but I don't see how the math works.

[guckyfan]

I'm just going to add a thought. Just because a stage it deemed "reusable" and launched it may not be "reliable"... One needs to factor this into the cost structure estimates...

IMO reusable implies reliable. An unreliable stage won't fly, SpaceX or any other provider cannot afford it.

[nadreck]

A stage costs no more than $15 million. We know this because expendable FH costs 30 million > expendable F9.

I don't see where you can infer that it is expendable first stages on the launch vehicles quoted at $62M and $90M on http://www.spacex.com/about/capabilities. The inference from the performance limits written in small print right below the price is that they are recoverable launches.  However, I do infer that an expendable F9 must cost more than $90M or there would be no incentive to use an FH.

[schaban]

I think what gets constantly overlooked are savings on logistics.
You save on whole operations between California, Texas and Florida for the most costly piece of equipment. Including tests.
You eliminate static fire and WDR.
those are significant parts of launch cost. much bigger then leasing ships

[Rocket Science]

I'm just going to add a thought. Just because a stage it deemed "reusable" and launched it may not be "reliable"... One needs to factor this into the cost structure estimates...

IMO reusable implies reliable. An unreliable stage won't fly, SpaceX or any other provider cannot afford it.

Yes I agree in theory, however the conclusion can not be drawn until repeated successful "reliable-reused" flights have been demonstrated...

[tp1024]

A stage costs no more than $15 million. We know this because expendable FH costs 30 million > expendable F9.

I don't see where you can infer that it is expendable first stages on the launch vehicles quoted at $62M and $90M on http://www.spacex.com/about/capabilities. The inference from the performance limits written in small print right below the price is that they are recoverable launches.  However, I do infer that an expendable F9 must cost more than $90M or there would be no incentive to use an FH.

If we assume that reusability reduces maximum payload by about 30% and take 22.2 tons to GTO as a baseline for FH, then a reusable FH should have a payload of just under 16 tons. FH could carry two 8ton payloads. If a single 8-ton-payload costs $90mio, they will probably try to get two payloads. (The market being what it is, they will probably be in the 6-7ton range.)

So, it is reasonable to assume a (fully expendable) launch price on the order of $180mio for FH. Triple the price of a fully expendable F9. And why shouldn't that be the case?

FH will require a special core stage that will not benefit from the same economies of scale as a regular F9, more quality control to ensure proper detachment of the boosters plus the necessary equipment. It does need dedicated erectors and launch pads that must somehow be financed. I would also consider it likely that FH will get a bigger upper stage than F9, a beefier payload adapter (the current one is only suitable up to 10.5 tons payload) and probably also a bigger fairing.

There will be some extra cost, so the price is quite justified. Frankly, $90mio for an FH launch without reuse strikes me as quite ludicrous.

[abaddon]

So, it is reasonable to assume a (fully expendable) launch price on the order of $180mio for FH. Triple the price of a fully expendable F9. And why shouldn't that be the case?

Only one S2, interstage, PLF.  Any non-hardware cost (range fees, payload processing, etc) are likely to be closer to F9 than triple F9 costs.

[nadreck]

If we assume that reusability reduces maximum payload by about 30% and take 22.2 tons to GTO as a baseline for FH, then a reusable FH should have a payload of just under 16 tons. FH could carry two 8ton payloads. If a single 8-ton-payload costs $90mio, they will probably try to get two payloads. (The market being what it is, they will probably be in the 6-7ton range.)

If you look at http://forum.nasaspaceflight.com/index.php?topic=39181.msg1521480#msg1521480 which was modeled with the original full thrust numbers you will see that this is not the case. Also why would they mix an LV cost for F9 with a shared payload cost for FH side by side?

You can also play with my spreadsheet model see http://forum.nasaspaceflight.com/index.php?topic=34077.msg1500227#msg1500227 and the posts it refers to for more details.

So, it is reasonable to assume a (fully expendable) launch price on the order of $180mio for FH. Triple the price of a fully expendable F9. And why shouldn't that be the case?

The side boosters have been described by Elon as almost identical to F9 cores and current estimates by a number of members on NSF (myself included) are that S1 cores cost between $20M and $25M to produce.

There will be some extra cost, so the price is quite justified. Frankly, $90mio for an FH launch without reuse strikes me as quite ludicrous.

Exactly my point $90M is for a launch with all cores recoverable. In fact with all cores RTLS'ing  (EDIT to add a parenthetic: "at least as far as my modelling suggests")

[Mader Levap]

A stage costs no more than $15 million. We know this because expendable FH costs 30 million > expendable F9.

Wrong. 60 milion is for expendable F9, but 90 milion is for at least partially reusable FH.

[oldAtlas_Eguy]

Let me see if I can clear uop some confusion that seems to run rampant on this subject.

The terms expendable and reusable when applied in the context of flight profiles does not drive costs. It is whether the booster per GS is new or used. This is because recovery is still a much higher risk than launch and basing costs on flight profiles would be at first backwards with a fully expendable actually costing less than a recovery. In fact it would always be that way. But a recovery gets you back a stage that can be reused that for the next launch would enable offering a discount.

The other confusion comes from the association that is made because the two items are next to each other of price and payload. The prices are for new boosters. The payload is for a recovery profile. The point being that with new boosters an expendable profile would have the same price.

Now for where that $90M price for FH came from. Back when the F9 was priced at $54M and there was two hardware versions for FH of crossfeed and no-crossfeed, the $90M number was associated with the no-crossfeed. A value of $125M was shown for the crossfeed. The key here is that the current makeup of the FH is a no-crossfeed and would follow closely to that pricing of $90M. But also that $90M was for new and not used boosters as well. Although reuse was on the horizon it was not part of the pricing workups at that time.

The conclusion is that with other statements of GS is that the price of a F9 with used booster would be ~$43M and a FH with all 3 used boosters probably ~$50M (~6M more because of the cost of refurbishment etc of the 2 extra boosters). In modeling the cost of manufacture the cost of a 1st stage of $17M makes all the number s come out right if SpaceX only makes the same total profit on an FH flight (lower profit margin) that they do on a F9 flight. Making all other costs and profit for a flight besides the cost of the 1st stgaes manufacture to be ~40M for either F9 or FH. So the differences in price for an F9 with used booster and a FH with 3 used boosters will be ~$6M more vs $30M more when both have all new boosters.

It that latest which will have the largest impact on the space industry.