In the thread Space X What am I missing:
https://www.rocketryforum.com/showthread.php?139192
sopirV wrote:
I'm sure this will be a lightning rod of a post, but can someone explain why the approach of the SpaceX team with falcon makes sense? The argument I understand is that significant costs can be saved if we don't modify the aeronautics of a reusable vehicle- IOW, vertical take off is best suited by vertical landing.
Where I'm stuck is the fact that the vehicle needs to carry/reserve fuel for the landing. How is that more economical than supplying enough of the good stuff to get the job done, and then having to save some we can return vertical?
I've moved my reply over to this thread as I think it is more appropriate, and others who follow this thread can see it. Some of what I mention here will be news (like Falcon 9 Block 5). And some a re-hash of discussions going way way back. But part of it I don't think was discussed in the same way, such as the way I'm starting it off below.
Let's say an expendable rocket's first stage needs to burn "100,000 units" of fuel by staging, for its second stage to put a given payload mass to a given orbit. Well, say you may need 6% of that fuel, or 6,000 units in order to be able to land on a barge on the ocean, and account for extra mass like landing legs and steering grid fins. So, you do not build the booster to hold 100,000 units of fuel, you build it a bit bigger (longer) to hold 106,000 units of fuel.
Why "only" 6% more fuel to be able to land? Because the rocket is VERY light with the upper stage and most of the fuel gone. So it does not need a lot of fuel to be able to maneuver, and land.
Yes, it costs a BIT more, but it is way less than 6% more cost since the main thing they need to do is to make the fuel tanks 6% longer, which does not cost much in raw materials. And if your company BUILDS its own tanks (as SpaceX does, they build as much as they can in-house), the labor and machinery is not going to cost much more to make the tanks 6% longer. The cost of the engines, guidance system, and so forth is the same (long as there is enough thrust, and the current uprated version of the 9 Merlin engines have plenty of thrust). As long as the extra mass does not over-stress something that should have been made a bit stronger (and indeed some components may need to be a bit stronger).
So, it costs a bit more to be able to have the extra fuel to be able to land on a barge in the ocean. If it lands successfully, then it can be a whole lot cheaper to be able to re-fly it, than to build a new booster. The unknowns, publicly, is what it costs for doing the ocean recoveries, what it costs to refurbish it to be able to fly again. But if the booster costs $30 million to build, and it costs $5 million total for recovery and refurbishment, that's $25 million savings. Now they have definitely spent many millions on the two ASDS landing barges, and some more on the dock infrastructures, but they also have spent many million on the launch pads, hangars, and other facilities. So the up-front cost of the ASDS barges should be amortized over time by the number of boosters that land on them that can be reused, which otherwise would splash into the ocean.
As things stand currently, it SEEMS like SpaceX is not giving a discount on previously used rockets, calling them "flight proven". Though nobody really knows for sure what the customer for the first re-flown booster will pay, full price or maybe a "secret" discount.
Now, I referred to using 6% fuel to be able to land on a barge at sea. Well, what if the customer payload is extra-heavy for the orbit it needs to go to, like GTO (Geosynchronous Transfer Orbit)? Make use of that 6% of "landing fuel", to get the payload into orbit, and not try to recover the booster. And that is exactly what will happen on the next flight, Echostar-23. It is so heavy that the booster will be expendable, it will not have landing legs or grid fins.
But let's flip that problem around the other way. What if the payload only needs to go to Low Earth Orbit, and/or is not very heavy? So there can be more fuel left over, like say 10%? Well, with 10% fuel left over, it can use that extra 4% to "boost back", literally stop its downrange motion and push back to Return To Launch Site (RTLS). That is exactly what CRS-10's booster did on Sunday. The Dragon spacecraft only goes into LEO, and is not THAT heavy (relative to the current Falcon's capability), so there is a lot of extra fuel available. So, it was able to use that extra fuel to land back at the Cape. Which saves time and money compare to sending a barge and a ship or two out to sea, spending days on the ocean and then back to port and using a crane to lift it from the barge to a special stand on the dock.
BTW - when Falcon Heavy (FH) is operational, it will have the greatest payload lifting capability since the Saturn-V. It will use three Falcon-9 boosters. All three firing at liftoff, then the middle throttling down a lot to use less fuel while the two side boosters do most of the work until the side boosters shut down. But the center core will have a LOT of fuel left in it, so after the side boosters are gone it can throttle up and fly a lot father downrange before it shuts down and the 2nd stage takes over. On a typical FH flight, the two side boosters will RTLS back to the Cape. And the center core will land on an ASDS Barge in the ocean. So, if things go well, they will get back all three of the first stage boosters. On some lighter payloads to LEO, they even plan to land the center core back at the Cape too. While for some really heavy payloads, they'll use all the fuel for the payload and the center core will be expendable.
I know this sounded extremely unlikely a few years ago. But they've developed this very carefully over the years, doing a number of landing test flights at McGregor, Texas. And they have worked out the bugs of the launch vehicle for landing more reliably (though "stuff happens", so if they get in say 10 flights I won't be surprised if there's a booster crash landing this year, I would be surprised if they crashed several). The last landing failure due to the vehicle was a bit over a year ago when a landing leg latch failed to lock, allowing the leg to fold and rocket to fall over. They also had a crash later but they knew it was a high-risk landing (extra-heavy payload so they had less fuel to use) and literally ran it out of fuel before it could land.
What's certainly left to be seen is how well the re-used boosters fare. SpaceX has already announced that late this year they will have "Block 5" of the Falcon-9, with some upgrades. And part of the upgrades are to make it capable of holding up to the re-entries better, so they expect to get many flights per booster out of that version. Which implies that the current ones won't get reused much, maybe only two flights, probably not more than three (A couple of the boosters that made risky super-hot re-entries, may never fly again. But the damage they had is showing SpaceX what they need to work on).
In late March or early April, there will be the first re-flown booster.
BTW - The above examples of 6% fuel for ASDS barge landings, and 10% for RTLS, are from memory of discussion on NSF forums 2 year ago or more. So, i'm not quite sure of the numbers. The RTLS might be 12%. And in any case it was "experts" doing the math to derive the % it would take. SpaceX has never said, IIRC. But in any case, even if a few percent off, the examples are useful for understanding how it does NOT require a lot of fuel to land, and that if a rocket is built on purpose to have more fuel than it needs to put a given payload into a given orbit, it can have the capability to use extra fuel to land vertically. And the way SpaceX builds their rocket boosters, it does not cost them much more to build longer tanks to achieve that capability.
Also, a lot of simplifications in the examples. For example, if the rocket needs 6% of fuel to land at sea, and you build the rocket to be able to hold 6% more fuel...... that is not enough. Because of the added mass of that extra 6% of fuel, and the added mass of the longer tanks. So...... it needs MORE fuel (and slightly longer tanks to hold that extra-extra fuel) in order to lift the extra 6% of fuel. Mind blown yet?
May not need more than 1% extra-extra fuel though (not sure).
So, yeah.
They have to account for all that kind of stuff too. But they did it.
Finally, below, is a drawing showing the various versions (Including the original single engine Falcon 1). The original Falcon-9 (v 1.0) was shorter than today's, and had no capability for landing. Then they stretched the tanks (v 1.1) to give the Falcon not only more fuel to land, but far more fuel to boost heavier payloads too. Notably they also stretched the upper stage tankage as well, a major performance boost. Then for v 1.2 they uprated the Merlin engines, calling it "Full Thrust" or Falcon FT (not officially calling it v 1.2, IIRC). They are inconsistent with their designation for the versions, as shown by "Block 5" coming up late this year, when they have not used the word "Block" for any previous version (and it is not clear if the current vehicle is therefore the 4th version of Falcon 9 or what, since they've only had 3 publicly designated versions. And FH is supposed to be "current version Falcon" with modifications as needed to adapt the current type to a Falcon Heavy core and side boosters. When block 5 comes around then FH will also use block 5 boosters, modified to fly as FH).
See Wiki at:
https://en.wikipedia.org/wiki/Falcon_(rocket_family)