I assume it burns up, but I’m having trouble finding the information. I know at one time they looked into returning that as well as the first stage for reuse, but I think they gave up on it. So, my assumption is that it’s got fuel left after separation of the Dragon capsule, then it essentially deorbits and burns up in the atmosphere. Right? Anyone know?
I think this Stackexchenge answer is still correct.
For a LEO mission, it is intentionally deorbited using its remaining fuel.
For a GTO mission the perigee is so low that it always decays quickly.
Cool, thanks for the info! I thought that was the case, but in a discussion with some friends this weekend about the up and coming Demo 2 flight one of my friends was skeptical. I told her I thought there would be fuel left to de-orbit the second stage, as they originally planned to actually try and recover them, and that was the most logical plan to keep additional space junk out of orbit, but also that I thought that it would decay anyway, regardless. She wasn’t buying it and for some reason my Google-Fu was weak and I couldn’t get a definitive answer that didn’t involve a dreaded video.
”Quickly” in this context, means only decades or hundreds of years. It is very difficult to loft a payload up above the lower part of Low Earth Orbit (LEO) and then have enough remaining impulse to shape the apses into a decaying orbit. Although it is desireable to deorbit all residual hardware that is not an operating satellite, SpaceX, like every other space launch operator, has left plenty of second stage structures (“rocket body” in satellite parlance) and debris in orbit as can be seen in Stuff In Space (set the group to SpaceX and search on Falcon). For this reason and for more precise targeting, many payloads use an apogee kick motor to shape the apses while the upper stage of the booster stack flies into a lower (eventually decaying) orbit, or for Medium Earth Orbit (MEO), Geosynchronous/Geostationary Earth Orbit (GEO), or High Earth Orbit (HEO) they’re often placed in specific disposal orbits, and for transplanetary injections they are often kicked into extremely long period orbits or outside the Earth’s sphere of influence entirely and go into a solar orbit.
In the case of CRS and CCP missions, there is a need to deorbit the stage so it does not pose a hazard either directly or as a result of collision, so there are special rules regarding the particular azimuth of the ISS. Even with that, the ISS has to perform evasive maneuver several times a year to move out of the way of debris that could potentially impact it. This is usually planned in conjunction with an orbit raising maneuver but occasionally some previously untracked piece of debris or change in orbital parameters of the debris results in an ‘emergency’ maneuver (one that is planned in days or a couple of weeks rather than the typical schedule of orbital maintenance operations).
Stranger
Very nice link!! Thanks Stranger!
I admit I thought they would arrange for those second stages to decay within months.
Unless I misunderstand, most do. I count only six items from 2019 on that Stuff in Space link, while there were thirteen missions.
Clicking each one, on the Stuff in Space site, I see a few with a very low perigee (less than 200km) but very high eccentricity. Apparently those have quite long lifetimes.
Here is a site about earth orbit lifetimes. Toward the bottom of the page, it seems you can indeed deduce that orbital lifetime goes up very rapidly with eccentricity, even for a low perigee.
To expand on that, high-eccentricity orbits mostly tend to decay by circularizing first. A decrease in speed at perigee (the closest point) will pull down the apogee, but will leave the perigee unchanged. Eventually, once the apogee is low enough, you’ll start getting decreases in speed well away from the perigee point, and that’s what finally starts lowering the perigee.
Unless, of course, your perigee is low enough to allow for lithobraking. Then you can get extremely rapid orbital decay.
On the other hand, it only takes a very low delta-V at apogee to make a huge difference to the perigee. So if you’re trying to be a responsible citizen of space and clean up your litter, it would take a very small maneuvering rocket fired at apogee to do that.
That’s how I thought it would work.
Although the Stackexchange link says GTO missions do not in practice perform that small deorbit manoever, because it would be too big a risk and liability.
Here is an odd case:
The Wiki for Asiasat 8 says the second stage GTO orbit decayed within one months from 195 km perigee to 185. Two months after that, it was 169 km.
Today it’s six years later and it is still listed on Stuff in Space as 2014-046B and on N2YO
When is that thing predicted to decay?
When the payload separates from the upper stage it is in the same orbit, so if the payload is a satellite in a fairly high and stable orbit, the amount of impulse to get the stage back down to a perigee that will encounter significant drag may be significant. Of course, at that point the stage no longer has to carry the payload and has consumed most of the propellent that it carried at launch, so you are getting more change in velocity (∆v or ‘delta-v’) per mass unit of propellant but it is often prohibitive to put it into a reentry orbit.
The orbital specific energy of an object is actually independent of eccentricity, depending only on the semi-major axis of the orbit (see the vis-viva equation and the second boxed equation here for a clear derivation). A more eccentric orbit will have a longer semi-major axis by definition, so it will take more time to lose energy aside form the fact that it only spends a small amount of its orbit in the lower thermosphere where some drag occurs.
A “very small maneuvering rocket” on the upper stage is still dead mass that has to be carried all the way up into orbit so it is a 1:1 trade with payload mass, and most launch systems are sufficiently marginal that the reduction in payload mass is often at the limit (including growth margin that you have to build into any mission). There is also the issue that if you have to rely on this retropropulsion system for deorbit requirements then it has to be build with the necessary redundancy to achieve the requisite reliability albeit probably reduced from the reliability requirement for the primary mission objective, and the possibility that if this system fails to correctly control and guide the rocket into the precise decay trajectory may place it somewhere that you don’t want it, e.g. on intercept with something else in orbit with a wide range of uncertainty, notwithstanding stability issues of trying to reverse a stage that may have sloshing residual propellant or other fluids.
For those reasons and more, we generally avoid designing additional unnecessary propulsion systems, either relying on the residual propellant to put the stage into a disposal trajectory of some kind or just planning the mission such that the range of orbits the stage ends up in do not threaten to intercept anything else. Anywhere above LEO, it is often just safer to push the stage into a somewhat higher orbit that has been designated for disposal (a so-called ‘graveyard orbit’) than to try to direct it slowly back down through other orbits where a collision may occur. Leaving an object in orbit may seem counterintuitive compared to forcing the stage to burn up upon reentry, but every propulsive operation carries both uncertainty and risk of malfunction whereas putting things in controlled orbits where they can be actively tracked means the probability of collision can be controlled down to a statistically acceptable threshold. Someday we’ll have to figure out how to collect all of that ‘junk’, which is a good project for non-military applications in space.
Stranger
Maybe this is interesting:
Compare AsiaSat 8 and AsiaSat 6.
Asiasat 6’s second stage decayed in less than 4 months. Asiasat 8’s second stage is still up there after almost 6 years.
Difference?
Asiasat 8’s payload mass was 100kg higher.
According to the press kits (PDF links), Asiasat 8 had a very slightly shorter second stage boost, very slightly longer coast and very slightly longer GTO injection burn.
All of which were apparently enough to leave it with just a slightly higher perigee and a LOT longer orbit lifetime.
So was this really due to the mission being at the limits of the launch vehicle performance? Are there rules requiring sufficient performance margins to guarantee a short enough second stage orbit lifetime?
It doesn’t take much of a difference in perigee to dramatically reduce drag. The thermosphere starts at 100 km (more or less) but the density is dependent upon the scale height, which itself is determined by the masses of the atomic and molecular species and the temperature of those species as they absorb energy from sunlight. Although the temperature of the thermosphere is often stated as an averaged value, because there are so few interactions between particles every species can actually have a different temperature dependent upon its absorption spectrum. So even just a few tens of kilometers can mean an order of magnitude difference in effective drag.
There is some general guidance in terms of how long, when, and where objects can persist in orbital space, but there is really no regulatory framework to enforce any rules aside from strict liability, e.g. if your debris smashes a telecom bird then you can be sued in court, and if you manage to accidentally take out a national security asset there will probably be some guys in suits and aviator sunglasses knocking on your door. Remember, too, that ever kilogram of payload that you take all the way to orbit is twenty or thirty additional kilograms of propellant on the first stage, notwithstanding the additional propellants needed on the second (and subsequent, if) stages, so you may actually be looking at a ratio of more like 100:1 for adding mass in addition to the payload and other required inert mass like the propulsion system, flight computer and other avionics, flight termination, sensors, payload deployment, payload fairing, et cetera. And of course you want to have a certain amount of margin to be able to make corrections in case the first stage trajectory is at the edge of the box, or you have lower than expected performance, or whatever, so also requiring enough margin to assuredly put the second stage into a reentry trajectory can easily become a mission driver or completely prohibitive.
Stranger
Thanks Stranger, very interesting.
I shouldn’t have said performance “margin”, I meant performance. The mission design data from the press kits was known and planned in advance. So the lower perigee must have been a conscious decision and compromise to get the heavier mission done on the launch vehicle they had.
Would they have accurately predicted and accepted the very long orbit lifetime in advance?
The lighter mission launched later. Is it reasonable speculation that they went to great lengths to sacrifice payload just to get enough performance for a more rapid orbit decay? Could they skimp 100kg on stationkeeping fuel and would that be a reasonable compromise for this purpose?
Correction “So the lower perigee must have been…” of course I meant to say “the higher perigee must have been…”
My guess is that’s due to solar cycles: The upper atmosphere expands and contracts due to solar activity. So if the atmosphere was expanded to that thing’s perigee at the time that it launched, it would have initially decayed quickly, but now that the atmosphere has contracted again, decay is slower (for now). If that’s it, then it’ll resume decaying quickly in about five years.
I doubt any significant changes were made to the primary mission parameters–and none whatsoever to the payload–to facilitate disposal of the second stage. Generally speaking, the payloader comes in with orbital insertion requirements and the launch vehicle provider figures out if and how that can be accomplished with disposal of the vehicle being a secondary (but still important from a liability standpoint) consideration. It would be a rare situation that a payloader would give up any mass margin just to satisfy some ancillary goal of the launch vehicle unless there was some mission reliability reason for doing so, e.g. assuring protection from post-separation recontact or contamination. The last thing a satellite with an onboard propulsion system is going to trade is propellent for maneuvering because the amount of propellant typically dictates the service lifetime of the spacecraft and that is often measured in months or even years per fraction of a kilogram (depending on propulsion system and maneuvering requirements). If the launch vehicle needs to put the upper stage in some disposal trajectory then it is up to the launch vehicle provider to figure out how to do that within the parameters of meeting the payload mass and orbital insertion requirements. That can often be done just by selecting an optimal (but narrower than otherwise) launch window, shaping the powered trajectory and coast period(s) to minimize propellant usage, or just offloading ballast if they can tolerate that and meet mass properties requirements for control.
As a rule of thumb, the payloader will start out about at about 60% to 70% of the launch vehicle payload mass capability at manifesting, will work up to 85% at system requirements review, 95% at preliminary design review, and ~105% at critical design review, and the launch vehicle has to figure out what they can remove to make up the shortfall before the flight readiness review. (That’s an old joke but it’s not far off; I worked one mission where the payload requirements grew so quickly they eventually determined that they needed a completely different launch vehicle, fortunately before PDR.) Now, the “payload capacity” is an advertised capability and the actual mass the vehicle can carry will depend significantly on the trajectory, notwithstanding that heritage launch vehicle operators will maintain a conservative mass reserve that they can provide when the payload grows past the nominal value; unfortunately, many of the nascent commercial providers provide what can charitably be referred to as “optimistic” capabilities that keep declining as they realize that the mass of the launch vehicle itself grows as it approaches design maturity. (Who knew that fasteners and wiring harnesses contributed mass even if they aren’t represented in the CAD model?)
But payloaders are almost never willing to make changes to accommodate the launch vehicle–which is just a stupid ‘truck’ that takes their precious payload into orbit where it will do magical things to amuse and astonish–unless they absolutely have to or there is some mission success improvement that makes it really worthwhile, and even then they whine about it to everyone who will listen. (Again, I kid, but I cannot count on the fingers of both hands the number of missions where the payloader and launch vehicle contractor have ended up deeply antagonistic with one another over the inability of one or both to meet mission requirements as defined by contract.) So, I strongly doubt that the payloader made any changes on AsiaSat-6 to facilitate post-boost stage disposal. Without looking at details of the payloads (though all of the recent AsiaSats were built by Space Systems/Loral using their LS-1300 common bus platform) I would guess that there are differences in the power system requirements that explain the 100 kg difference.
Stranger
Thanks again Stranger. So fascinating. I love posts like this!
Thanks all for answering the question. Don’t know if anyone was following live, but they just aborted the mission for today due to weather. Not sure when the next planned alternative is. I assume they will have to unfuel the rocket and probably do a bunch of stuff for maintenance and prep to get it ready for the next try. Kind of disappointing, but safety first, especially for this mission which really needs to go well with lives on the line.
Next launch window is on Saturday.