NASA's Europa Clipper leaves for Jupiter's moon atop Falcon Heavy SpaceX has sent NASA's Europa Clipper on its mission to the Jupiter moon atop its Falcon Heavy rocket. Falcon Heavy ready to launch Europa Clipper (pic: NASA) Falcon Heavy ready to launch Europa Clipper (click to enlarge) – Pic: NASA It was the sixth and final flight for the first stage side boosters, which have been used …
"It's pretty incredible that we're living in an age where they defend sacrificing a rocket booster to complete a mission."
It's costly in performance to recover rocket stages so for a mission that must have all of the umph the system can give, it's appropriate to splash a booster or three. All of the missions that SpaceX flies where they do land the booster for reuse means the F9 is more rocket than needed to do the job. Maybe it would be cheaper to have an "F6" or a Falcon/Raptor, or ..... Starship is bigger as there isn't the volume capacity in the Falcon series to launch Starlink satellites 60 at a go that would be far more economical. With the later versions of the satellites, 20 at a time is about all the can get except for some extreme trajectories.
Falcon heavy payloads are few and far between. The plan is to get Starship costs way below Falcon 9. That would take away any possible payload for a five core Falcon.
Five core Falcon would probably burn out two cores as fast as possible and send them back to the launch site as they would be low on propellant before the get very far (like a Falcon heavy). The next two cores would burn a little longer and likely land on drone ships. By the time the center core gets low on propellant it would be even faster and further away than a Falcon Heavy - which is unlikely to have its center core return to drone ship. The Falcon 9 fairing is a bit small for a Falcon heavy. There is an extended fairing but it will fly so rarely that SpaceX are not going to develop recovery for it.
It might be fun to calculate if a five core Falcon could get Orion and its service module to NRHO. SLS lobbyist would go incandescent with rage.
"I wonder if SpaceX has considered using 4 boosters instead of 2?"
It's a mass vs. volume thing. To qualify a larger fairing for bigger payloads would be very expensive and might not lead to enough business to make the effort worthwhile which leads to no need to have more mass capability or higher orbital insertions. The controls get exponentially more complicated the more boosters that are strapped on.
For sure, Elon admitted that he though FH would be a couple of months of development instead of the 3 years it ended up taking. He found out that its not as simple as strapping on a couple of boosters. He was even talking about cross-flow between boosters and centre core before realising there wasn't much benefit vs effort it would take. Even just the ground changes for 4 boosters would be huge, fuelling 5 boosters at once would be interesting.
"For sure, Elon admitted that he though FH would be a couple of months of development instead of the 3 years it ended up taking."
It has to be remembered that Elon is not an engineer. He's mostly a super-charged "influencer". It's been some time since Tom Mueller left SpaceX and he was the principal architect of the Falcon rockets. He didn't want anything to do with Starship which might be a big reason he left.
You should Everyday Astronaut's interviews it you don't think he has no engineering knowledge. Also Eric Berger's books Liftoff and Reentry are definitely worth a reading.
And of course NSF Flame Trench podcast on YouTube, not engineers just geeks that have been following the space industry for 15 years. They aren't SpaceX fan bios and do give them a hard time when needed.
Not to downplay Tom Mueller's part but his background and contributions main contributions were propulsive, obviously instrumental in the companies success but certainly not the principle architect of Falcon rockets. He was lead developers for Kestrel, Merlin and Draco. He was also head of propulsion when Raptor concept goes back to 2008 when it was actually going to replace Merlin on Falcon upper stage. Serious development started in 2018 and Mueller has said that he worked on Raptor from 2013 until he left in 2020, so a bit hard to state that he had nothing to do with Starship, He attributed leaving mainly burn out. I suspect working for Musk for so long will do that to you even if you love your work. If you Youtube him, there are some pretty good in depth interviews, really interesting guy. Maybe you know something different ?
As for Falcon design, you had early employees like Chris Thompson and Hans Koenigsmann who had big parts to play in their own areas. Probably the biggest advances that Falcon had was in Software and there was several teams working on that.
Try costing it out. The internal cost of an extra Falcon 9 launch is $18M-$25M depending on who you ask. I will go with $20M. A full load of propellant is $½M so not much room to save money there. For a first approximation propellant and materials scale with payload. In real life some things do not scale for example guidance, navigation, communications, license and operation. I will approximate a Falcon 5 as half the performance of a Falcon 9 but 5/9 of the scalable costs. An expended Falcon 9 launch to GTO can carry 8300kg so expended Falcon 5 would be 4150kg. Falcon 5 expended has to compete on cost against Falcon 9 return to drone ship (5500kg to GTO). Throwing a Falcon 9 booster into the sea costs SpaceX about $27M so a Falcon 5 booster would be about $15M. That leaves $5M for the upper stage and non-flight hardware costs. A Rocket Lab electron launch has a price of $7.5M. Their profit margin is thin and their complete rocket is much smaller a Falcon 5 upper stage would be. That makes an expended Falcon 5 launch a similar cost to Falcon 9 return to drone ship.
Now lets strangle it. Falcon 9 costs $300M to develop over Falcon 1 so Falcon 9 got engine development for free. Falcon 5 needs $300M development cost plus a new upper stage engine. Economies of scale mean Falcon 9 takes all the Starlink launches leaving only crumbs for Falcon 5. Falcon 5 has to repay its development cost and ground support infrastructure over a small fraction of the number of launches. Worse than that, it takes a few launches away from Falcon 9 so its yearly costs get divided by a slightly smaller number.
"All of the missions that SpaceX flies where they do land the booster for reuse means the F9 is more rocket than needed to do the job. Maybe it would be cheaper to have an "F6""
No - it means there is enough performance to land. It would in no way be cheaper to throw away a fractionally smaller rocket.
"No - it means there is enough performance to land. It would in no way be cheaper to throw away a fractionally smaller rocket."
It does. By fractionally, if you mean 10% smaller, yeah, no sense in that. Something that's 50-60% of a F9 might be cheaper even as an expendable system. Rockets are mostly fuel tanks and they don't scale in a linear manner. The deciding factor is customers. I'll discount Starlink as even Elon has come out and said profitability in the long term using F9 to launch Starlink sats doesn't look good. They need to loft more per mission and need the ability to build bigger ones with more/different functionality. It's also an in-house program. If payloads are small and lighter as time goes on, the F9 winds up being too big.
It's not just fuel to land either, It's the mass of the landing legs, increased margin (more mass) in the engines and structure. Needing a barge for retrieval if there's a need for cross range boosting. There's all of the infrastructure to inspect and refurbish in addition to dockside handling for barge landed boosters.
The work I was doing was on landers rather than boosters, but in a room full of rocket nerds from all sorts of places, we discussed these things and landing, of course, was a big part of what we were doing. Precision landing was a big part and this was all before Elon/SpaceX started their Grasshopper.
"It does. By fractionally, if you mean 10% smaller, yeah, no sense in that. Something that's 50-60% of a F9 might be cheaper even as an expendable system. Rockets are mostly fuel tanks and they don't scale in a linear manner. "
No they scale horribly - for your concept to work they'd need to scale worse than linearly, and they don't, bigger rockets have significant advantages.
The Electron, for comparison, can do about 320kg to LEO for about $8m
An F9 can put about 23 tons to orbit (expended), 17.8t (Downrange), probably 13.5 RLTS
It costs ~$70m (with a recovered booster)
That's 18t/320kg - well over 50 times the payload for only ten times the price.
And the vehicle is a little more than three times the size (i.e. 8 times the volume).
The engines are expensive (over 50% of the cost of an F9 booster), and previously they've been thrown away every launch.
SpaceX have just expended a booster which has made 23 flights - that's about $400 million in savings from not having to build new engines.
Every recovery they make is worth ~$20m in engines alone, probably in excess of $35m given the rest of the booster even accounting for maritime costs and refurb.
Doesn't matter how you cut it - you'd need to make a booster that was the cost of a single merlin engine to come close to beating reuse.
The chipmunks say that Falcon 9 upper stage kept running for 1/2 a second after being commanded to shut down, resulting in an overshoot of the recovery area.
Apparently a valve failed, so a backup valve was used after the computer noticed the thrust wasn't tailing off.
People are all "oh it's the used engines breaking" and I have remind them the upper stages are expendable and it's a new engine each flight.
I find it amazing that after getting all the way to Mars, 12% of the distance to Jupiter, it is worth coming back to Earth for a Mario Kart jump boost. The ESA Juice mission is going all the way in to Venus on the way to Jupiter. These space boffins are incredible. It must be much more difficult to calculate the trip when the goal is to drop into orbit when you get there, rather than snapping pictures on the way past. Travel times to Jupiter:
Voyager 1 fly past: 546 days
Voyager 2 fly past: 688 days
New Horizons fly past: 405 days
Juno into orbit: 1,796 days
Europa Clipper into orbit: ~2000 days
I wonder if the launch delay means they have to redo all of their calculations. I would say that's annoying but I expect they love their jobs.
"coming back to Earth for a Mario Kart jump boost."
Nice analogy :-)
I wonder when we'll see a mission that, for whatever reasons, they'd like to get there a bit quicker and it can pause in orbit, re-fill the tanks from a handy orbiting tanker, and blast off in the right direction? All these slingshots a very clever, but it can be very, very slow. Can a Falcon heavy reach orbit with the main core still in place? Or is that asking too much of the launch system?
"Can a Falcon heavy reach orbit with the main core still in place? Or is that asking too much of the launch system?"
Not even close... the FH core is about 2/3ds depleted when the boosters are dropped, and it's at about 5-10% of orbital velocity...
Actually looking up some more data... I might have underestimated the extra oomph they get:
https://www.reddit.com/r/SpaceXLounge/comments/1347tx0/today_the_core_of_falcon_heavy_set_the_record_of/
That's a year ago, but reckons that fully expended the central core can get to about 17Mm/h, which is closer to 20% of orbital velocity.
That second stage does an *awful* lot of the velocity.
"Not even close... the FH core is about 2/3ds depleted when the boosters are dropped, and it's at about 5-10% of orbital velocity..."
Thanks! So, at best, there might be the opportunity to refuel a 2nd stage once it's in orbit. Possibly not worth it, unless someone wants a fast trip to the Moon :-)
"redo all of their calculations"
Your comment made me think of the movie Hidden Figures. I saw it for the first time about a month ago. Brilliant film, I'd recommend anyone that hasn't seen it yet (i.e. me a month ago), and has an interest in the early space program watch it.
These missions start in orbit, then they wait for the right moment in orbit to boost to the new orbit, or out of earth orbit to a sun centric orbit etc.
Redoing all their calculations means rerunning through the computers and checking them. Start with Ansys AGI STK and work up (forget Kerbal).
Getting into orbit around Jupiter is a lot trickier and more expensive than just flying past it.
Also, because Europa happens to be in a region of heavy radiation thanks to Jupiter's magnetic field, the orbit of the spacecraft has to be a pretty extreme ellipse to reduce the amount of time it's in that region to avoid cooking the electronics and the solar panels.
This was explained in one of the NASA videos on the mission.
re calculations
suspect one or more of
1) they do the calculations for several dates so they can move to an alternative date with ease
2) the flight is re-calculated once in initial Earth orbit
3) one or two days difference at the start makes little difference overall and the trajectory is adjusted in flight anayway
If I recall correctly, If they could have used an SLS to launch the Europa Clipper it would have cut the transit time significantly since it is a more powerful rocket. But there aren't any spare SLS rockets right now so the time advantage is lost in waiting, and a Falcon Heavy expended is far, far cheaper than an SLS.
"If I recall correctly, If they could have used an SLS to launch the Europa Clipper it would have cut the transit time significantly since it is a more powerful rocket. "
What ever energy you build up getting there has to be countered to get into orbit. If you just want to fly by, it doesn't matter. If New Horizons was going to orbit Pluto, it would have taken ages to get there while carrying enough fuel to slow down as well.
Ion engines can be rather useful since they can run a long time. Accelerate halfway there, turn around, and accelerate in the opposite direction the second half of the journey (more or less). I recall seeing some simulations on going to Saturn and by the time the craft hit the halfway point, it would be going like stink even with a rather low ∆V. I expect that engine reliability and fuel management will be super important.
Its around 2 years difference between SLS and F9. When you factor in costs, you have think that you now need 2 years extra mission support (maybe even longer) plus do you need to redesign any of the space craft to cater for the longer duration ? So actual cost may not end up being that different. Its a mute point, there are no spare SLS available for next 10 years and test data indicated that Europa Clipper would need redesigning for cater for extra vibrations for a SLS launch though that has been widely meet with scepticism.
There is a big difference between doing a flyby vs trying to orbit. All that energy you spent building up velocity to get there has to be countered for orbital insertion, with flyby you can go as fast as possible. Also relative distances between Earth and Jupiter aren't ideal with the slingshots they are now having to do. Calculations are easy enough but when your shielding was expected to last x amount of time and now needs to last an extra couple of years, plus power and you have staff mission control and all the other bits. So yes its cheaper to launch on FH but in terms of overall mission costs it might not be cheaper.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
