Slashdot Mirror
Estimating SpaceX's Reusable Rocket Cost Savings (theverge.com)
An anonymous reader writes: On Monday, SpaceX successfully landed its Falcon 9 rocket after launching a group of satellites into orbit. It's a huge breakthrough for the commercial space industry, because reusing rockets will dramatically reduce launch costs. The question now is: by how much? Elon Musk says it takes $60 million to build the Falcon 9, and $200,000 to fuel it. That's a big difference, but we can't expect them to immediately launch the rocket again after refueling it.
"The Falcon 9 experiences major temperature changes during its flights, as well as intense pressures and vibrations from the winds in the atmosphere. These all produce wear-and-tear on the vehicle's hardware — meaning the rocket might need repairs and updates before it can launch again." This kind of refurbishing is why the Space Shuttle ended up being way more expensive than expected. Fortunately, the Falcon 9 is not nearly as complex.
This is now the true test of SpaceX's design talents; if the rocket is built to be durable, then repairs and replacements could keep relaunch costs very low indeed. Steve Poulus, a former NASA project manager, suspects final costs could be driven below a million dollars. That figure would give SpaceX the capability of easily underbidding any competitor for government contracts, not to mention bringing it into affordability for any number of companies who'd like to put a satellite in orbit.
"The Falcon 9 experiences major temperature changes during its flights, as well as intense pressures and vibrations from the winds in the atmosphere. These all produce wear-and-tear on the vehicle's hardware — meaning the rocket might need repairs and updates before it can launch again." This kind of refurbishing is why the Space Shuttle ended up being way more expensive than expected. Fortunately, the Falcon 9 is not nearly as complex.
This is now the true test of SpaceX's design talents; if the rocket is built to be durable, then repairs and replacements could keep relaunch costs very low indeed. Steve Poulus, a former NASA project manager, suspects final costs could be driven below a million dollars. That figure would give SpaceX the capability of easily underbidding any competitor for government contracts, not to mention bringing it into affordability for any number of companies who'd like to put a satellite in orbit.
The wear mechanisms are poorly understood and this compounds the difficulty of predicting the reliability of launches. Only time will tell if the Falcon technology can operate more reliably than the historic 1-2% failure rate. I am not optimistic. A 1% catastrophic failure rate is nothing to boast about, even for a reusable launcher.
1) The first stage is 2/3 the total cost to launch. Which would be $40 million.
2) They can renovate the first stage for $5 million.
3) They can get five launches from a first stage (original plus four more).
So, $60 million for five launches, plus the $20 million for the second stage x5.
Which comes to $32 million per launch. A bit more than half the current price.
Now, I consider those pessimistic assumptions.
Alternately...
If we replace (3) with 15 launches per first stage, we get $28M per.
If we replace (2) with $1M per launch, we get $29M for five launches, $23.75 per launch for 15 launches.
Big picture: reusing the first stage only will allow them to drop prices by 40-60%.
Now, if they can reuse the second stage also, we're talking some real money....
"I do not agree with what you say, but I will defend to the death your right to say it"
The wear mechanisms are poorly understood, and this compounds the launch risk. After decades of engineering work, heavy lift boosters still have about a 1% failure rate, If SpaceX can't beat that, then there is no glorious future for manned spacefight.
Elon Musk and company have been making huge achievements and seems one of the few people in industry to take the long view of things and it's likely to pay off in the end even though the MBAs hate him at the moment.
I, for one, welcome our new Martian overloard!
"Be particularly skeptical when presented with evidence confirming what you already believe." -
The 60 million cost includes the cost of manufacturing the components, inspecting them thoroughly, assembling them and inspecting and passing the assembly. They went for proven technologies and cheap materials, says the article. The rocket shell is relatively simple. The cost is in the rocket motor, the nozzle that holds the combustion products in plasma state, its cooling mechanism, pumps and controls. I am not sure how much of it can be inspected and re validated without extensive disassembly. I suspect most of the cost is in the inspection and certification of the vehicle rather than in the cost of individual components or the cost of assembling them.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Still, there are obstacles. SpaceX still needs to demonstrate the ability to consistently produce and launch rockets many times a year after the June accident caused an unexpected, six-month setback, something it will do with several flights planned for the weeks ahead.
Just because it's relatively cheap to use Space X, if I have a 50-50 ( better or worse) chance that my $100 million satellite that took several years to design and build is going to get blown up, I'll pass.
I fully expect that they'll have a tiered price structure. First, the cost of a brand-new launch won't have to be $60,000,000+ if they expect to launch the rocket again. Obviously it'll be the most expensive and will probably see the most use for the most critical launches and for manned-launches once they're man-rated, but if they expect to launch the same rocket assembly a half-dozen times then that launch might cost $20,000,000 or $30,000,000, or less than half of the cost of a one-flight rocket. Limited-reuse rockets might fetch $10,000,000 to $20,000,000 dollars for important but not absolutely critical launches where budget is important but so are timetables, and less than $10,000,000 for launches on older rockets where the loss of payload isn't that big of a deal, like replacing GPS satellites or dealing with routine communications satellite replacement over time.
If this comes to pass, even if a customer insists on brand-new rockets for their launches, so long as there are customers interested in budget launches, the first launches will get cheaper.
Do not look into laser with remaining eye.
A quote from an NBC News story about Blue Origin, a company founded by Amazon CEO Jeff Bezos, shows one of the differences between SpaceX rockets, which go into orbit, and Blue Origin rockets, which don't orbit:
... important to clear up the difference between 'space' and 'orbit'," Musk posted on Twitter. ... A rocket needs to be traveling about three times the speed of sound, or Mach 3, to reach space, but orbital missions require speeds about Mach 30, Musk said.
SpaceX is working to reuse rockets that are returning from the higher altitudes and faster speeds of orbital missions. "It is
1) The first stage is 2/3 the total cost to launch. Which would be $40 million.
2) They can renovate the first stage for $5 million.
3) They can get five launches from a first stage (original plus four more).
So, $60 million for five launches, plus the $20 million for the second stage x5.
Which comes to $32 million per launch. A bit more than half the current price.
Now, I consider those pessimistic assumptions.
Interesting numbers. Let's try a variant case. Suppose in addition:
You're assuming that the non-reusable launch vehicle cost per launch is $60M. OK, let's start out by assuming 1/3 of that is fixed costs and operations costs, and 2/3 the vehicle cost, which is split evenly between the two stages (first stage is larger, but not proportionately more expensive). So, of the $60 million, $40 million is spent even if the vehicle first stage was free.
Now assume that re-usability increases the launch cost by, say, $5 million (launch operations are expensive! and the cost is not entirely the vehicle).
Assume that all the stuff needed to make the first stage reusable increases the stage cost by 25%, from $20M to $25M.
And assume that the delta-V and the added mass to do the fly-back decreases payload by 10%, and that the price you sell the launch for decreases a similar percentage (some payloads won't care, but some will.)
economics are much less clear now. The first stage cost drops with refurbishment from $20M to $6M, but the total launch cost only drops from $60M to $51M, whereas the price you can sell the launch for drops from $60M to $54M.
Still an economic advantage... but only a few percent advantage.
Satellites don't inherently have to be that expensive, the only reason they are is that you can't afford to launch another one when the one you have fails. Because launches are so expensive it makes sense to sink money into making the sat as reliable and long living as possible. If launches were free you could whip up a functional comms sat for thousands or tens of thousand and if one fails just lob another one. The cost of payload depends on launch cost, with half the launch cost you can make the sat half as reliable, for half the cost and just lob two of them.
"This kind of refurbishing is why the Space Shuttle ended up being way more expensive than expected"
It was fully expected to be that expensive, the upper management simply ignored it.
At one point when they were still considering fully reusable designs, the Phase II' candidates, management put the cost of the system at something like $100 per pound to orbit. However, they had already estimated the staffing at the Cape to be on the order of 25,000 people, which meant the payroll alone was about $500 a pound. Most estimates put the absolute lowest cost at $1000/pound. After Challenger it was over $2500, making it the most expensive launch system in US history.
So why was management saying $100 until the end? Because the entire justification for the Shuttle was that it would be lower cost than any other system. And because of that, everyone would move their cargos to it. And since everyone moved their cargos to it, it would be launching all the time. And because it was launching all the time, the embedded payroll per launch was lower. Even then it didn't look like it could match Scout, so they came up with the Getaway Specials to try to take those, and then cancelled Scout.
Now it was clear to everyone, including the very detailed CBO report, that if they didn't get every single payload out there, then there was no way to get the launch rates they needed to make the payroll costs go down. And as the CBO report noted, if any of those assumptions failed, it would end up being more expensive than systems like Titan. And they went on to point out that many of the payloads NASA assumed would move to the Shuttle never even existed in the first place (modular telcomsats for instance, which they just made up).
So management lied, fully aware there was no way they could meet the numbers. And it was this precise attitude that caused the Challenger Disaster, where bad news numbers were simply ignored and replaced with ones that met political or economic criterion.
Just because it's relatively cheap to use Space X, if I have a 50-50 ( better or worse) chance that my $100 million satellite that took several years to design and build is going to get blown up, I'll pass.
SpaceX has launched 20 times, with one failure. So, that's a 5% chance of loss, not 50%, assuming they maintain this success rate.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
The space shuttle program ended up being extremely expensive compared to rocket launches.
The Shuttle was the poster child for how NOT to do reusable. Government spec with 10 mission requirements orthogonal to each other, half of which are not technically possible at time of design while development is spread out over ever ZIP code in the country.
Elon can't be dumber than Congress. Unpossible.
Faster! Faster! Faster would be better!
[Citation Needed]
Rockets are not coins. So far, the failure rate seems to be 5%. If the one failure was a statistically early event and/or the issue around that one failure is fixed, the true failure rate will be much lower.
We need a larger data set to firmly set the real failure rate, but there is no evidence of 50/50.
In fact, if a per launch odds really were 50/50, the probability of 19 successful launches would be 1/(2^n) or 0.00000190734 or 0.000190734%
TLDR; you have no idea what you're talking about. Take a stats class.