Congratulations SpaceX for the 50th successful mission!
Pushing 6100kg in orbit and -in theory- being able to recover the booster is impressive. At least at first sight. Looking at this into more detail, there are major financial drawbacks in this mission.
Switching the mission to recovery came at a hughe cost I think. The orbit they achived is by far the lowest to date:https://twitter.com/planet4589/status/9 ... 0887444480
189 x 22257 km x 27.0 deg
Someone on reddit/spacex estimated this orbit takes an additional 400m/s to circularize compared to the other orbits. For comparison: the 6.7t intelsat 5E reached an otbit of 42748 km x 303 km x 25.8°. As Hispasat was lighter, a fully expendable mission would have achieved an orbit over 50.000km.
If this was booked expendable - and I think it would have been 3 years ago- , it would have had a pricetag of $95m https://twitter.com/elonmusk/status/963094533830426624
Stationkeeping for a geo satellite is about 50m/s per year, so 400m/s lost due to additional circularization would be 8 years less lifetime? The satellite was probably build with more deltaV, as a 2015 falcon 9 was not supposed to reach supersync orbits with such a payload, but probably not off by that much. Intelsat 35E was contracted to reach about 28.000km, if I remember correctly.
It is obvious that the customer got a huge discount for switching to expendable. Probably not the full $30million difference to the reusable F9, but as a customer, I wouldn`t take a >4year lifetime hit for under $20m( my guess)
So this unrecovered recoverable mission attempt was probably VERY costy for spacex. They had the full one-week cost for their recovery fleet on duty without any return. They lost expensive titanium gridfins and landing legs.
Lot`s of people asked elsewhere why they didn`t remove the gritfins. The answer is obvious. Last-minute changes are a recipie for desaster. Without gritfins, they also had to change software config from recovery to expendable with totally different parameters. And there are a lot of possible errors to make in the process last minute. Not a good idea. Going expended would have probably been impossible due to the defined notam/exclusion zone definitions too. The booster would have landed way further downrange than it did in recoverable mode.
Worst case this mission came out $32m less than anticipated, and minus one F9 block 4 in stock, which may have been included in the 30 launch plan this year.
To get the recovery concept working, they must work on the contract side to allow delays due to weather in case it is launchable, but not recoverable. They probably are now, but it looks like this older contract wasn`t or the short-term switch wasn`t well thought out contract wise and dates in it didn`t allow for a week delay for recoverable weather and avoid this loss.
I think this 50th mission isn`t something Spacex have cheered about financially.