Looking at the Cocoa florida facility, there are ~150 cars there.
Assuming 100K per employee, that is $15M in total cost to SpaceX.
Elon has previously claimed Raptors can be made at $200K per.
Steel of the grade of rolls we have seen going into SS retails for $1M/SS or so.
Three starships per year looks conservative as to what that facility may be able to make as a raw shell, if they do not also have to cope with initial learning, or facility construction.
This would put the fixed costs of a basic SS at a rather ridiculous $7M or so per. ($15M/3+$1.2M+$1M)
By basic here, I mean one with no aerofeatures at all and little else other than the bare skin, tanks, engines, and RCS.
Performance would be slightly better than nominal due to needing no aerofeatures, so ~180 tons, with 55 tons dry mass is probably plausible.
Nominal cargo Mars is something like SS takes off, is fully refuelled in ~LEO, burns for Mars, aeroenters with 150 tons of payload.
If we assume seven launches is this baseline scenario, what would be needed to equal this, with a vehicle incapable of aeroentry?
The harshest cargo scenario would be for example a launch in Jul 2020. This has an 8 month transit time, and enters at 5.6km/s.
Assuming no aerodynamic effects, a pure propulsive landing with 150 tons requires the tanks to be full at aeroentry.
Injection from LEO is another 3.6km/s, so if we are dealing with tanker/depots moving only when filled, we need somewhat under 4* the total mass at mars entry in LEO - 4000 tons.
Or, 22 launches.
Neglecting superheavy for the moment, this puts the total cost of cargo landed on Mars at ~$1000/kg, if we assume the cost per SS is actually $7M.
It puts the cost of cargo on the moon at under $300/kg. (Assuming ~5500m/s)
And cargo in a capture orbit around Mars below even that.
Very significant improvements for Mars surface (halving the cost) can be had if the vehicle can cope with a mach 9 aeroentry, as Elon has stated SH can manage without any special coatings.
There are issues I have omitted on the negative side - boiloff - but I have also neglected production speeding up as serial production gets underway.
I note that at the time of BFS unveiling, the cost of a CF SS was widely assumed to be of the same order as the above 20 'disposable' starships, meaning the economics of this would be basically the same as that case for the first couple of synods until SS starts coming back in numbers.
And of course - once you have serial production of starships ongoing, reentry trials with the tankers you're throwing up there just became really cheap.
Anyone else with Program Management experience want to express an opinion here?
Lots of program management experience, none of it in aerospace.
But I think that misses the OP's very valid point which is that this is a multi-order of magnitude paradigm shift. Quadruple all his costs if you want; pretend it costs $28 million to send a ship to Mars.
Looking at the Cocoa florida facility, there are ~150 cars there.
Assuming 100K per employee, that is $15M in total cost to SpaceX.
Elon has previously claimed Raptors can be made at $200K per.
Steel of the grade of rolls we have seen going into SS retails for $1M/SS or so.
Three starships per year looks conservative as to what that facility may be able to make as a raw shell, if they do not also have to cope with initial learning, or facility construction.
This would put the fixed costs of a basic SS at a rather ridiculous $7M or so per. ($15M/3+$1.2M+$1M)
By basic here, I mean one with no aerofeatures at all and little else other than the bare skin, tanks, engines, and RCS.
Performance would be slightly better than nominal due to needing no aerofeatures, so ~180 tons, with 55 tons dry mass is probably plausible.
Nominal cargo Mars is something like SS takes off, is fully refuelled in ~LEO, burns for Mars, aeroenters with 150 tons of payload.
If we assume seven launches is this baseline scenario, what would be needed to equal this, with a vehicle incapable of aeroentry?
The harshest cargo scenario would be for example a launch in Jul 2020. This has an 8 month transit time, and enters at 5.6km/s.
Assuming no aerodynamic effects, a pure propulsive landing with 150 tons requires the tanks to be full at aeroentry.
Injection from LEO is another 3.6km/s, so if we are dealing with tanker/depots moving only when filled, we need somewhat under 4* the total mass at mars entry in LEO - 4000 tons.
Or, 22 launches.
Neglecting superheavy for the moment, this puts the total cost of cargo landed on Mars at ~$1000/kg, if we assume the cost per SS is actually $7M.
It puts the cost of cargo on the moon at under $300/kg. (Assuming ~5500m/s)
And cargo in a capture orbit around Mars below even that.
Very significant improvements for Mars surface (halving the cost) can be had if the vehicle can cope with a mach 9 aeroentry, as Elon has stated SH can manage without any special coatings.
There are issues I have omitted on the negative side - boiloff - but I have also neglected production speeding up as serial production gets underway.
I note that at the time of BFS unveiling, the cost of a CF SS was widely assumed to be of the same order as the above 20 'disposable' starships, meaning the economics of this would be basically the same as that case for the first couple of synods until SS starts coming back in numbers.
And of course - once you have serial production of starships ongoing, reentry trials with the tankers you're throwing up there just became really cheap.
If it's unreasonably cheap, SpaceX will raise the price to customers until it's "reasonable" and pocket the profit for their own use.
I think the point of Starship is that it’s unreasonably cheap. When it’s fully reusable with no refurbishment, launch costs drop to basically fuel plus ground support. Like, kilobucks instead of megabucks or gigabucks. So cheap that launch cost and mass are no longer a constraint. What the heck do we do with that?
I think the point of Starship is that it’s unreasonably cheap. When it’s fully reusable with no refurbishment, launch costs drop to basically fuel plus ground support. Like, kilobucks instead of megabucks or gigabucks. So cheap that launch cost and mass are no longer a constraint. What the heck do we do with that?
If it's unreasonably cheap, SpaceX will raise the price to customers until it's "reasonable" and pocket the profit for their own use.
They have already set a bar with F9 launch costs, so for similar payloads that could fly on F9, the customer costs will likely be similar (or slightly lower to incentivize launching on a newer vehicle) to that of F9. For payloads that could not launch on any existing vehicle or that want to be the only payload launched, they could charge a dimensional premium or something (for the opportunity cost of not being able to rideshare).
I think prices will be set at a level that is the highest it can be while still making it nonsensical for customers to launch with a competitor rather than SpaceX. The incentive to lower it will come through competition as other players develop similar technologies. In the meantime, SpaceX would do well to figure out how to further minimize launch and manufacturing costs so by the time the industry catches up, they can cost-effectively reduce their pricing further. Good ol' competitive dynamics!
In other words if SpaceX can launch their spaceship for... let's just say $10 million, and still make ends meet, why would I spend $20 billion developing my spaceship 2.0? If I can only make $2 million margin on each of my future launches then I'd have to launch it 100,000 times just to earn back my initial investment.
I think there's a good chance prices will go much lower that Falcon 9, as SpaceX tries to grow the launch market. Making a lot on each launch is good, but making a bit on many, many launches is even better.
We've already seen this happen with their smallsat rideshare program: they started at $2M per ride with one launch per year, then cut the price in half and added a bunch of launches when demand was high.
I think there's a good chance prices will go much lower that Falcon 9, as SpaceX tries to grow the launch market. Making a lot on each launch is good, but making a bit on many, many launches is even better.
We've already seen this happen with their smallsat rideshare program: they started at $2M per ride with one launch per year, then cut the price in half and added a bunch of launches when demand was high.
In the scenario posited above, SpaceX decide to go all-in on inexpensive starship before getting reusability working, as it saves financially over F9 launches, and allows interesting projects like Mars at the same cost as was baselined in 2017.
They have a baseline vehicle constructable at a rate of dozens a year with a skilled workforce.
Once they nail reusability, and can with most of the same skilled workers and facilities pivot over to reuse, not expend, things for other vendors go from terrible to utterly reliant on governmental and corporate restrictions on not launching on spacex.