Hacker News new | ask | show | jobs
by aaron_m04 2007 days ago
It just isn't plausible that Starship launch costs for LEO could be 1/28th the costs for Falcon Heavy. This sounds like more Musk optimism to me.
5 comments

Falcon Heavy has a significantly smaller payload bay and the 1st stage will always be lost. If a starship can fly 10 times and requires much less refurb than a FH, it makes sense.
Right, the projected rate is effectively the super-bulk discounted rate if you buy 10 launches worth of cargo at once and actually fill it to max capacity whenever a launch window is open. I don’t think the individual flight rate would be anywhere close to $50/kg for the vast majority of customers.

The more relevant number is the all-in immediate price per launch, the rate where one could simply wire the money and SpaceX would launch next week, divided by the actual thing(s) launched, which will weigh some intermediate amount. And compare that to the same for the other rockets.

Nobody does launches on a week's notice.
Right, in practice prices in the launch industry are not ‘prices’ as a consumer would understand them because any quote comes with certain contractual restrictive covenants, agreements, timelines, etc., that really are more akin to home buying than a typical commodity priced in per kg terms.

For order of magnitude comparisons we can ignore all that, for finer grained comparisons it really is comparing apples and oranges.

In the ideal future world launches should be done without the need for secret contracts, something like the private jet industry where planes can chartered, almost at the push of a button, at short notice.

Soon we won’t remember when SpaceX wasn’t launching every week.
Launching every week is still very different from launching on a week's notice.
Saying it just isn't plausible that ambitious thing will happen without an argument for why it won't happen... doesn't really lead to an interesting conversation. Especially when all the arguments for why it is possible already exist in public forums.

I mildly disagree with you, but I can't even begin to take the other side of the argument without knowing why you think the price estimates given aren't plausible (or even basic facts like what the cost you are using for the Falcon Heavy is...). Other responses to you spew out random facts trying to argue that Falcon Heavy is expensive in ways that Starship isn't, but I really can't tell if those random facts are even closely related to why you think it's implausible.

The reason I thought it was implausible is because in my experience, such dramatic drops in price never happen in only a few years.

The other responses did help clarify how it can occur with Starship.

Part of the problem is that it's not quite that high, and it's also a moving metric. Apparently the cost of a Falcon 9 launch is down to $28 million, which is ~$1,200/kg thanks to reusability.

″[The rocket] costs $28 million to launch it, that’s with everything,” Couluris said, adding that reusing the rockets is what is “bringing the price down.”

https://www.cnbc.com/2020/04/16/elon-musk-spacex-falcon-9-ro...

As SpaceX uses their Falcon 9s more, the cost will continue to fall and approach consumables plus the amortized cost of the 1st stage.

The Starship is the same idea, but with larger payload and better reusability. Realistically it'll be somewhat expensive initially, but as they improve the design and reuse, costs will drop.

They are designing for full and rapid reusability (like an airplane) and unlike the Falcon Heavy, the second stage is not thrown away, but also fully reusable. Seems plausible to me. If I remember correctly Musk estimated $2 million cost per launch, with about 1/4 of that raw fuel costs.
The upper stage of an F9 or FH costs $10m and is expended every time. The fairings also cost $6m ($3m per half fairing) and are pretty tricky to recover. No part of Starship will be expended every time, it’s all reusable. The only marginal cost per flight is the fuel and operations costs. They’re not going to reach 1/28x with the early models, maybe only 1/10, but eventually they’ve got a reasonable chance.