The news that Amazon founder Jeff Bezos is now publicly committing to using the United Launch Alliance’s (ULA) giant Atlas 5 rocket was no great surprise. However, his decision allows a noticeably clear opportunity for a ‘contrast and compare’ between the Bezos-backed Project Kuiper and arch-rival Elon Musk’s Starlink scheme.
The two rivals are also the two richest men on the planet, so a billion dollars invested here or there is just small change to them.
Choosing the Atlas V rocket is probably a wise decision given its ultra-reliable record of launches. According to the ULA Atlas V has a 15-year launch proven record of successes and a heritage of more than 600 launches.
Musk cannot quite manage that 15-year heritage, but the success rate of his Falcon 9 rocket is totally unrivalled and it has one spectacular extra benefit: his rockets come back to Earth. This reusability has always been key to the Musk formula. Bringing the rockets back (and including the nose cone fairings) saves big money on each and every launch.
An Atlas V launch costs a minimum of $109 million. It is not yet known what version of the Atlas family Bezos has chosen but, even with the basic Atlas V 401 variant, it means the possibility of lifting more than 8000 kgs to a low Earth orbit (LEO) position. A 401 with a ‘medium sized’ payload (4.7m diameter and 12m-13m cargo bay), and has the potential for many, many satellites carried into orbit.
Musk normally charges around $60 million for a commercial launch of a Falcon 9 rocket. SpaceX’s Starlink’s can be configured into a 60-satellite constellation stack, but using a recovered rocket booster stage (and perhaps also the fairings) slashes the actual costs bottom line to the Starlink project to a fraction of that $60-ish million.
In fairness to Bezos and his team’s negotiating power it is quite likely that Project Kuiper has enjoyed a discount on its 9 launch contract. How much might that saving be is anyone’s guess (other than the contract lawyers). It is known (from an EMusk Tweet last year) that his Falcon 9’s cost $1 million less just on insurance premiums alone.
However, what is undoubtedly exciting for potential users is that – at last – Bezos has firmly pressed the ‘Go’ button on Project Kuiper. The scheme has to start launching satellites, of course. Kuiper is licensed to launch in five distinct phases, the first being 578 satellites to a 630 kms height. Kuiper must launch at least half of its licensed constellation by July 2026. That’s more than 1600 satellites out of the 3236 total.
Musk has demonstrated that two launches of his Falcon 9’s per month means that this year alone 1440 Starlink satellites can be lofted this year. If Bezos can match that launch cadence for the next two years then real competition will be in place.
Then, it will all depend on price, service, and the usual metrics for any competing businesses.