Excellent analysis, Fred.

Reusing only the rocket engines is like an airline reusing only the jet engines after each flight, and the proportional cost for jetliner airframes is similar to the 80% for rockets, as mentioned in the article (I believe that it is about 70% for twin engine airliners — each jet engine being comparable in cost to the rest of the airframe).

Can you imagine how little business the airlines would have if the cost to fly passengers and cargo incorporated the delay and cost of installing engines on new airframes after each flight? Air travel might be as uncommon as rocket launches to orbit.

Refurbishment costs and schedule are important unknowns. NASA got an unpleasant surprise when they discovered that the Space Shuttle turnaround cost more and took longer than expected. SpaceX and the other private companies have an advantage over NASA on this point: they are motivated to find better designs and methods, over time, to reduce costs and schedule time — even if they are not so unpleasantly surprised with their current designs.

That is part of the competitive process. Faster, better, cheaper! Whether SpaceX succeeds or not, they certainly have driven the rest of the industry to be more efficient and reduce costs. Peter Diamandis, with his Ansari X-Prize, intended for the space industry to reuse hardware with short turnaround times. Space-X took up that challenge for orbital rocketry, and the rest of the industry is playing catch-up.

You are right about the cost of fuel. An extra 10% to 30% fuel consumption is cheap, compared to the cost of the rest of the rocket’s structure. As I understand it, the cost of fuel and oxidizer for an entire launch is less than $1 million ($700,000 for Falcon 9?), but the first stage, without the engines and avionics, costs well over $1 million (20% of somewhat less than Falcon 9’s $60 million basic launch price tag?).

Little more than what a high school kid could do.

By the time they finalize the design, start production on the whole new system, test it at least once they will have blown past any perceived savings in subsequent launches.
When has a fully government funded, designed and built project ever come in under budget and on time? Plus this is the Eu and they are having trouble with their old projects let alone a new one.
They will spend months trying to decide who gets the bolt contracts, then they will move on to the screws.

I think that the reason Ariane and ULA designers seem to prefer “engine only” recovery is they are stuck in the false “optimize payload” mindset. It is true that for a given size rocket, you can realize more orbited payload if you don’t fly back your entire first stage. Optimizing payload to orbit in this sense means spending minimum mass on recovery. Indeed, expendable rockets are fully optimized, spending every kg on sending mass upward and zero on recovery.

But this is an un-useful quantity to optimize. It doesn’t matter in the long run if a recoverable booster is twice as big as an existing booster to loft the same payload. Fuel is cheap.

The quantities to optimize for are:
* Total Cost
— Cost Efficiency of recovery / refurbishment
— Cost of operation
— Cost of manufacture

* Reliability
* Suitability to deliver the payloads of paying customers

Please, Blair, no apologies are necessary. Landing the first stage is obviously the most difficult part. From the business perspective of SpaceX’s competitors, however, getting that first stage capable of reusability is more important. That both ULA and Airbus are taking a different approach to saving the first stage, trying to save just the engines and avionics in a more controlled manner, suggests to me that the engineers at both companies have doubts that the Falcon 9 first stage engine will be as reusable as SpaceX would like, after landing.

Apologies. I’d thought the standard for ‘close’ was landing the stage. That is the most difficult part of the exercise. I agree that there’s a world of difference between ‘recovery’ and ‘reuse’, but in only two tries they’ve come damn close to making the first assault on the problem, while not solving it entirely. I’m assuming reuse is engineered into the manufacture of the stage, so if they can get it down reliably, the rest should be relatively straightforward.

They have come close to landing the first stage, but that isn’t yet close to reusing it. After landing they will have to do additional tests to see what shape it is in, and whether refurbishing will be required to make it reliable enough for another commercial flight.

That seems a bit uncharitable. I’d say by any reasonable standard, Space X has indeed ‘come close’.