What can we learn from SES-9?

SpaceX are launching SES-9 next week [1], and this launch has the potential to tell us a great deal about the capabilities of the new Falcon 9 Full Thrust. Two weeks ago, SpaceX agreed to loft the satellite into a higher orbit than previously planned [2], at the cost of a lower probability of first stage recovery*. This makes the launch very revealing, since a trade-off like that implies the launch will operating right at the limit of the rocket’s capacity.

One thing to note immediately is that, according to SpaceX’s own website [3], the Falcon 9 should not be able to launch SES-9. They claim a payload of 4850 kg to geostationary transfer orbit (GTO), while SES-9 weighs 5300 kg [1]. One caveat to that is that SpaceX are probably not launching to a standard GTO, but to whatever orbit the Falcon 9 can muster. Even so (neglecting this effect), it’s possible to estimate what the equivalent payload capacity to low earth orbit (LEO) would be if 5300 kg is the new (approximate) correct maximum to GTO.

If you assume all of the increase to 5300 kg payload comes from better first stage performance, you get an upgraded payload capacity of 14,440 kg. If you assume instead the full increase in payload comes from having more propellant in the second stage, this much propellant would get you 14,370 kg to LEO, and if you assume it’s all from other improvements in the second stage, you get 13,950 kg. Of course, the reality is that the improvement come from a difficult-to-estimate mix of these, but they paint a pretty consistent picture – if 5,300 kg is an accurate measure of SpaceX’s GTO ability, then expect around one extra metric ton from their LEO capability.

Now the launch next week may tell us (among other things) if SpaceX really can get 5300kg to GTO and recover the rocket. This is particularly significant, because (as reflected by the meaningful probability of failure [2]) not even SpaceX know if this will work.

There a bunch of possibilities that could happen on launch day:

  1. The launch is successful, but the rocket falls short of the barge.
  2. The launch is successful, and the rocket makes it to the barge, but makes a hard landing.
  3. The launch and landing are both successful.
  4. Some kind of launch failure. God forbid.

Now what would these scenarios tell us about the capabilities of the rocket?

If the rocket falls short of the barge, it could tell us that 5300 kg really is above F9’s capacity. An alternative possibility would be that the conditions were just unfavorable that day – but how much will these really matter in terms of how much fuel is needed to reach the barge? We can do some estimates to see.

Although the rocket does three burns to slow itself down for landing, it also relies on drag to slow itself down without using up fuel. In the upper atmosphere above the Atlantic, wind can be much stronger than on the ground, on the scale of 100m/s [5] [6]. Keeping in mind that the rocket separates at around 1700 m/s [7], and must slow down all the way to 0 m/s, this could make a significant difference: the drag force is about 20% greater at 1000 m/s with a 100 m/s headwind than with no wind (because drag scales as the square of speed), all other things being equal. We don’t know the exact degree to which drag matters, or to what extent SpaceX will use forecasts to avoid launching into a tailwind, or which part of the atmosphere contributes the most to drag, but the numbers here seem to suggest that the weather could make a significant difference to SpaceX’s chances of landing the rocket. Of course, wind is not the only weather effect to make a difference – air density also affects drag (more density gives linearly more drag) and air pressure affects the efficiency of rocket engines (lower pressure gives more efficiency), although these are likely to be lesser effects.

If the rocket lands hard on the barge, we will probably see another explosion. But is it plausible, if the rocket runs out of fuel just as it lands, that we will see a crash, but no explosion?

Let’s run the numbers. The first stage carries an estimated 123,100 kg [8] of liquid kerosene. The energy density of kerosene is 46 MJ/kg [9], and one tonne of TNT releases 4.2 GJ [10]; in other words, 0.1% of the fuel (40 m/s of propulsion) is an explosion like this:

So there will pretty much always be an explosion, but the fireball size might tell us something about how much fuel is left. After all, we have plenty of previous fireballs to compare to:





So a successful or hard barge landing will help confirm to us that SpaceX has slightly more capacity than they claim, a hard barge landing might also tell us roughly how much margin they had spare, while failing to reach the barge could be as simple as running into unfortunate winds.

We wish SpaceX the best of luck with the launch, and the landing attempt.


The launch has now occurred, and we got option 2: satellite deployment was successful, while the first stage made it to the barge, but crashed. Making it to the barge is certainly a partial success, but sadly (as of three weeks after the launch) good quality footage of the crash has not been released, so we can’t easily estimate the speed it hit the barge, or the fuel remaining.


*They’re doing this so that SES will have to spend about month less time using ion thrusters to get the rest of the way to the target orbit; this is partly in compensation for the seven months of delay or so caused in part by the CRS-7 failure.




[1] https://spacexstats.com/missions/ses-9

[2] http://spaceflightnow.com/2016/02/09/ses-says-spacex-will-launch-its-satellite-in-late-february/

[3] http://www.spacex.com/falcon9

[4] http://spaceflight101.com/spacerockets/falcon-9-v1-1-f9r/

[5] http://skepticalscience.com/jetstream-guide.html

[6] http://www.netweather.tv/index.cgi?action=jetstream;sess=

[7] https://youtu.be/ivdKRJzl6y0?t=24m20s Jason-3 (a v1.1 rocket) was chosen over Orbcomm-2 (the only full-thrust launch) because it was a barge landing, whereas Orbcomm was a land landing.

[8] http://spaceflight101.com/spacerockets/falcon-9-ft/

[9] https://en.wikipedia.org/wiki/Energy_density

[10] https://en.wikipedia.org/wiki/TNT_equivalent




One thought on “What can we learn from SES-9?

  1. I think you’re missing a possible outcome which seems most likely to me. Chances are they’ll put the barge in the right spot for a landing but they’ll need to use more atmospheric breaking, which could result in upper atmosphere failure of the stage from pushing thermal and preasur boundaries of the stage.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s