Every single aspect of space travel is an engineering/mathematic/scientific marvel. Not only did we plan, build, launch these, (before I was born) but we're still communicating (until we can't).
I'm reading links people have posted here, trying to understand how we communicate with these probes. It's fascinating.
If you look at old spacecraft hardware, one thing that stands out is its apparent simplicity and down-to-earth (no pun intended) design --- and I'd argue that this is at least partially responsible for its extreme reliability.
From that perspective, I feel as though developments in modern technology just can't compete for impact; we constantly search for new ways of designing things, wrapping ourselves in endless layers of abstraction and high-level thought, yet aren't really "getting off the ground" and accomplishing something concrete, so to speak.
I’m not really sure this is true, though I understand why it might feel that way sometimes.
I’m currently travelling at about 180mph on board a high-speed train in Japan. I flew here on a jet which is something like 20% more efficient than the equivalent from a few years ago. Using the ubiquitous LTE network, I can make a real-time HD video call to my family back in the UK, using my palm-sized, battery-powered computer. I used the same device earlier to do some research about cities as we passed through them, and also to check the CCTV system at home. Over the past couple of weeks I’ve used a similar technology stack to locate my position to meter-level accuracy, to read and translate foreign language text from images in real time, and to record hours of 4K video.
Modern technology is astonishingly powerful - and in some ways, the examples I described above are even more impactful to me on a day-to-day basis than space exploration is. Don’t get me wrong - the latter is still important and exciting! But it’s sometimes too easy to forget the impact of the somewhat more mundane technology that’s all around us.
Which is also (high speed trains) an 30+ year old technology, even if the train you're on was built more recently.
That's what the parent means about those older technologies having more impact. Which is trivially true: earlier low hanging fruits give substantially more bang for the back to progress, and then you get incremental progress and finally marginal returns on any technological fields.
The airplane was a huge development. The modern commercial airliner (50+ years old by now) as well. A jet "which is something like 20% more efficient than the equivalent from a few years ago"? Not so much.
>Using the ubiquitous LTE network, I can make a real-time HD video call to my family back in the UK, using my palm-sized, battery-powered computer.
Which again, compared to the initial impact of the internet and mobile communications it's just an incremental improvement. Being able to send messages and talk from Japan to the UK instantly -- great impact. Being able to send HD video on top of that? Not so much.
Something built back then and still works.
Edit: fix markdown syntax.
Consumer electronics from a few decades are not quite cellphones, but not quite high speed trains or nuclear reactors or space rockets. Many old C64 systems still work or can be restored, and I bet most of our current high end laptops will continue to work a decade from now (you might need to replace the battery).
The OP might have been talking about efficiency, and we have gotten a bit sloppy with that in the consumer world (why does Slack/Atom/Discord need to be a 100MB+ app bundled with its entire web browser and framework? It's like we're in the 2000s with 15 copies of the JDK on your system again!), but once again .. different uses.
A modern SpaceX craft is going to have custom real time operating systems designed specifically to preform much more complex calculations than we've done in previous space missions, hopefully increasing reliability and the amount of sensors we can read, record and transmit data for. The software engineers might be less space efficient in their code than the previous generation, but if the hardware is cheaper and we can increase readability at the expense of memory, why not do it?
In Kim Stanley Robinson's Mars trilogy (highly recommend; best Sci-Fi I've ever read), humans eventually create AI so complex it can manage space factories designed to build from asteroids. The most advanced AI ever created is used to maneuver an asteroid into orbit of Mars while also mining the interior and constructing the cable that would eventually turn into the space elevator over the course of a decade.
Electronics aside, voyagers nuclear energy supply fascinates me.
I'm still very optimistic that someday we'll figure out safe micro nuclear reactors.
The energy density of nuclear fuel is just amazing.
But we've definitely made progress on a number of things. My rav4 although a lot more complex than Toyotas of 30 years back is a lot safer and efficient.
AI hype notwithstanding, no machine learning comes even close -- it's generally a parameter search in a space that's too pedestrian, leaving the real hard work (defining the problem context, goals / objective function, viable tradeoffs and shortcuts) to the human.
Cute AI demos aside, when reliability comes knocking on the door, you end up looking for ways to simplify or avoid the whole bloody mess.
If you spend the time to understand the problem well enough (as you must with space tech), the number of degrees of freedom aka model parameters shrinks. Then suddenly computing power and large-scale parameter searches don't buy you as much; their trade-offs against increased complexity aren't as appealing.
"With four parameters I can fit an elephant, and with five I can make him wiggle his trunk." -- John von Neumann
The only way we can even contact Voyager is the fact that we have better radio signal processing on earth and that has been improving since we launched Voyager.
Yes, but then they weren't using it to power immense social networks with pictures of people's food.
Just curious but does anyone know it's mission in interstellar space? Is it more than just a fun experiment at this point?
tl;dr: it's not clear what defines the edge of the solar system, there's interesting reasons to want to know where it is, and voyager taught us a lot about it.
For those in the UK, it's currently streaming on iPlayer - though it should ideally be seen in cinemas to be fully appreciated.
> "The Voyager flight team dug up decades-old data and examined the software that was coded in an outdated assembler language, to make sure we could safely test the thrusters," said Jones, chief engineer at JPL.
I understand the sentiment, but your phone was not made to exit the solar system.
Cellphones are a mess because we can't even have a nice base hardware platform. ARM isn't a platform. It's a SoC spec with random shit soldered to random pins by different vendors with completely non-upstreamable kernels. Google could just mandate UEFI on OHA phones like Microsoft did with theirs, but instead we're just getting this /vendor partition in the next release.
I don't think it's unintentional either. It's an aspect of planned obsolescence. The cellphone industry wants you to upgrade every two years, when we should not be destroying the planet and creating gear that lasts 10 years. Fewer factories, less pollution, longer life .. but we're in a consumerist economy hardwired the opposite direction, where any type of profit shortfall or lack of growth is seen as a problem, not the result of a good product.
What we need is a privately held cellphone company that will forgo profit in favour of creating long lasting, repairable, maintainable devices. [I've been working on this thesis for the transition from capitalism to communism]
Meanwhile I'm wearing a 25 year old tshirt, whilst tshirts bought much more recently wear out and get holes in.
The problem is the lack of vendor participation in the community.
Generally speaking, we built at least 3 versions (4 versions depending how you count).
1) Flight. The one put in the payload.
2) Flight spare. Often used in one of the 2 or 3 Mars test facilities at (the 2 Mars yards or simulation facility) and hooked up to a spare Ground Data System (controller system) via a 6 inch thick, 100 foot long “umbilical cord”.
3) A mechanical flight spare. A usually 80-90% mechanically accurate version that we called Bubba (at least for MER) that we’d pull out for Open House, or other special events and displays.
4) Lastly, I personally worked with subsystem flight spares (44u rack of the Satellite Communication Subsystem or the Flight Control System). Way too big to fit in a spacecraft in this configuration or a mock-up of the system.
What might be a more interesting story is how we bought dozens and dozens of old SUN pizza boxes and Sparc Stations in the mid and late 2000s from eBay (literally) in order to be able maintain the Flight Ground Data System.
AFAIK, the last update was in April 2010, “to delay turning off the AP Branch 2 backup heater for one year.“ (https://en.wikipedia.org/wiki/Voyager_2#Interstellar_mission)
Don't you just hate it when they make devices rely on cloud services... :P
"It is a difficult computer task to calculate what stars might by chance be along the Voyager spacecraft trajectories 50,000 or 100,000 years from now. Mike Helton of the Jet Propulsion Laboratory has attempted to make such a calculation. He calls attention in particular to an obscure star called AC+79 3888, which is now in the constellation of Ursa Minor -- the Little Bear, or Little Dipper. It is now seventeen light-years from the Sun. But in 40,000 years it will by chance be within three light-years of the Sun, closer than Alpha Centauri is to us now. Within that period, Voyager 1 will come within 1.7 light-years of AC+79 3888, and Voyager 2 within 1.1 light-years. Two other candidate stars are DM+21 652 in the constellation Taurus and AC-24 2833 183 in the constellation Sagittarius. However, neither Voyager 1 nor Voyager 2 will come as close to these stars as to AC+79 3888.
"Our ability to detect planetary systems around other stars is at present extremely limited, although it is rapidly improving. Some preliminary evidence suggest that there are one or more planets of about the mass of Jupiter and Saturn orbiting Barnard's star, and general theoretical considerations suggest that planets ought to be a frequent component of most such stars.
"If future studies of AC+79 3888 demonstrate that it indeed has a planetary system, then we might wish to do something to beat the odds set by the haunting and dreadful emptiness of space -- the near certainty that, left to themselves, neither Voyager spacecraft would ever plummet into the planet-rich interior of another solar system. For it might be possible -- after the Voyager scientific missions are completed -- to make one final firing of the onboard rocket propulsion system and redirect the the spacecraft as closely as we possibly can so that they will make a true encounter with AC+79 3888. If such a maneuver can be effected, then some 60,000 years from now one or two tiny hurtling messengers from the strange and distant planet Earth may penetrate into the planetary system of AC+79 3888."
We know so much more about exoplanets today than we did in Sagan's time, and have so much more computing power to bring to bear. Knowing the trajectory thrusters still work, it would be a fitting tribute to try one last interstellar bank shot into the corner pocket, and see if we couldn't honor Sagan's last wishes, and give the Voyagers a destination worthy of their journey and their cargo.
If so and Voyager only has enough power to do some minor rotations of the probe for three more years it's unlikely that there is enough power to actually change its overall trajectory, even if fired all at once.
BTW thanks for your comment, it was nice to hear Carl Sagan again :)
Looking at humanity today .. I'm thinking we'll go extinct. Love your loved ones. Don't spent too much time in the office. Life is too short to not really live, cause there's a good chance literally no one will remember us a million years from now.
The fact that ohur president in 2017 constantly throws out threats of nuclear war on a mass communication platform and yet has a sizable support says humans are fundamentally flawed. Just needs a few bad actors at the top and we'd be over.
Just watching those kids quantum stream their lifes in the Andromeda, yet none of them even mentioned the good old Voyager 1.
Certain Volkswagen models apparently have an even more amazing MTBF. It's a real sleeper. 
But Voyager's got it beat on mileage...
The idea here is that any radio energy that does not end up in the vicinity of the target was wasted and at 21 billion Km that gives you plenty of opportunity for mis-alignment.
Receiving the signal has similar challenges, with the added complication that this time the sender is sending with a power level that puts its signal under the noise floor by the time it reaches Earth.
Fun fact: this goes for the GPS satellites as well by the time their signal reaches your pretty little hand-held receiver and it takes nothing short of magic (to me, not to the people that design that stuff) to recover the signal.
For missions where that makes sense (mars probes, STS, ISS and IIRC even original Apollo moon landings) relay satellites or even networks of them are/were used. To some extent for such constructions to be useful it has to be constructed of satellites that orbit something which is near to target of the probe, which is impractical for probes that are on highly eliptical orbits around sun, not to say probes that are on exit trajectory like Voyager.
Edit: TLDR: for the amount of money required to design, build and deploy such relay satellite in geosynchronous or such orbit you can do several manned missions to mars or some other planet of your choosing
As for your edit: very heavy satellites cost (including launch) ~$250M whereas a manned Mars mission is estimated to cost $6B.
I really don't see how you could do 'several manned missions to Mars or another planet of your choosing' for the same budget as a single relay satellite, even a large one.
The problems I see with such a design are simply that it does not give you any advantage for spending all that money and actually stands a fair chance of making things worse. More stuff, so more stuff that can go wrong and if it does it is in a place where you can't fix it. Limited life-span as well compared to the Voyager itself because of the larger complexity (must keep two antenna's aimed at the same time while moving itself).
Uplink path of DSN is capable of significantly higher power levels than 20kW, several orders of magnitude more. Also the whole transceiver electronics are especially fiddly, with various cryogenically cooled and/or high-power microwave valves without meaningful solid-state replacements, which is not something you want to have in GSO without any chance of mainteance.
Yes, that's exactly what I wrote upthread, that's the hard requirement. And that is what will fail and then you've got a very expensive doorstop in an orbit outside of any repair capability. Inability do do maintenance / repairs is the killer.
On the other hand it is certainly true that RCS of such satellite would require exceedingly large stores of RCS supplies which will invariably run out and have to be somehow replenished.
Edit: even hall thrusters require stores of xenon and with the precision required the amount consumed is far from practical.
The Voyagers have a 3.7m diameter parabolic radio dish, larger than the Hubble space telescope's mirror even. That alone provides a huge amount of gain on communications. Additionally, the spacecraft have 10s of watts of power available for transmitting signals, which is a fair bit considering (while on the other end the ground stations have up to hundreds of thousands of watts to transmit). The ground-stations in the deep space network (DSN) are tens of meters across, a small antenna is 34m, the biggest ones are 70m across. That also provides a huge amount of gain alone. It means that there is more area to collect signals from the spacecraft and it means that the beam from the ground station to the spacecraft is much tighter, concentrating the total transmission power into a smaller cross-sectional area at the distance of the spacecraft.
The spacecraft also uses error correcting codes, which involve transmitting many more bits than the underlying data, but in such a way that errors due to noise are not only detectable but correctable.
On top of all of that you have the state of the art low noise amplifiers in the DSN antennae. A typical low noise amplifier is a carefully built electronics assembly made by experts. The DSN amplifiers? They use 99.95% purity ruby rods chilled to 4 degrees above absolute zero to form microwave MASER based amplifiers.
There's a neat little video (series) here on the DSN and contacting the Voyagers: https://www.youtube.com/watch?v=FzRP1qdwPKw
"Low-Noise Systems in the Deep Space Network"
Edited by Macgregor S. Reid
It's published by the JPL as part of the "Deep Space Communictions and Navigation Series". The rest of the books in the series, listed in the book's front matter, have some fascinating titles.
They mentioned that the received signal from the Voyager spacecraft is actually stronger than the signals from several closer craft, because the Voyagers have such good antennas.
The videos of their decoder screen brings back memories of doing very similar things with oilfield tools. The same sorts of techniques are used to get data from deep below the earth, though not with RF but mud pulse telemetry instead. Same digital encoding types and decoders. I got to work with the guys who designed all of the telemetry systems and wrote the decoders for that stuff too.
Curiously, those codes are now superseded by other, more modern approaches, for example Turbo Codes  which are used not only in deep space probes, but also in cellular communications and other applications that we consider normal these days.
So, starting from a system that can communicate from Earth to the moon, if you can find a way to add 64 decibels then it can work from Earth to Voyager. Ways to add decibels include using more directional antennas on one or both ends or transmitting with more power. Alternatively, you can make up some of those decibels by communicating much slower.
The interesting thing about the inverse square law is that it's insensitive to the scales involved. For instance, going from 10 meters to 20 meters results in a 6db loss, and going from 1 light year to 2 light years also results in a 6db loss. This is much different from, say, light in a fiber optic cable, which would experience a 6db loss from impurities in the glass each time the light traveled some constant distance.
If you could send a perfectly parallel beam, it would effectively be an antenna with infinite gain. As far as I know, that's not possible but getting as close as you can is a good strategy. There's also antenna aiming limitations to consider -- it's possible to have too much gain if it exceeds your ability to point in the right direction.
The opposite extreme is an isotropic radiator, which emits equally in all directions. (That isn't possible either, but it's a good theoretical baseline.) Antenna gain is usually described relative to an isotropic radiator. So, an antenna with a gain of 12dbi means that in the direction it sends its strongest beam, it's 12 decibels stronger than it would be if the antenna were an isotropic radiator.
I think that's the crucial point. While transmission through a cable etc. has exponential decay, transmission through vacuum has quadratic decay, so much more feasible.
The more interesting part is that Voyager can send signals we can receive (with giant, huge radio dishes, but still).
Keep in mind that there are inefficiencies in simply transmitting with more power. The more you amplify, the more noise you introduce. No matter what you do, you can never improve the size and hardware that you're transmitting to.
I don't work much with RF, but, with optical transmissions, the major innovations I've seen are from better receivers that are better able to separate signal from noise at lower power levels. On the earth side of things, we can use massive dishes connected to modern hardware that have very advanced signal processing capabilities. Ultimately, Voyager is 1960s-era hardware with very very very minimal ability to change the software in any way.
If they wanted to (and had the funding to) build a dish on earth that was 10x larger to receive the signals from Voyager with hugely advanced signal processing, that's a totally doable thing. On the other hand, there is virtually nothing you can do to make voyager hear better.
Ultimately, its a lot easier to amplify something faint than it is to shout louder.
Yes, but since you are amplifying the noise right along with the signal if the other side shouts louder it really helps. As does a very good directional antenna (parabolic, very solid mount, very precise control of its orientation).
(Searched for "voyager communication" on DDG, top hit.)
I was surprised to learn that it's a ground-based system. I'd think they would need antennas in space (on satellites) so that you can both send extremely powerful signals without disturbing others, and receive without having to go through the atmosphere. Instead, there are just three ground stations at approximately 120° around the earth for continuous communication.
NASA has a remarkble group of engineers who know how to get every last erg of energy out of that machine.
New plan for the evening...
Not with Voyager, but for example:
The year is 2243; a spacecraft that humans sent out of the solar system on a one way trip launched in 2045, has once again come into contact unexpectedly and is re-entering our solar system. The system seems to have been updated and is sending information that would have not existed when it was launched and is potentially foreign tech in 2243.
This ship was never supposed to come back...
Edit: Ok, I have admittedly not seen the first Star Trek movie.. I probably should =p
Rather belatedly the human race realizes - after decoding some interstellar video - that the Voyager's trajectory will bring it sooner or later into contact with a race whose sole mission is to eradicate all other intelligent life.
The Voyager's creators helpfully adding a map indicating its origins prompt the recall of the century: launch a mission to overtake both Voyagers and to capture them and bring them back before they are discovered.
Working title: Recall
(And you needn't bother, honestly. It's the franchise's answer to 2001, put together by people who didn't really get what made 2001 great. If, like me, you can't get enough absurdly prolonged sequences of old-school practical model effects in effectively static poses, then you'll love it to pieces, but otherwise...)
You just described the plot of the first Star Trek film. Even better, it was a Voyager probe (a fictional Voyager 6) that was augmented by alien machine intelligence and returned to Earth.
> alien machine intelligence
(Me, I like Ron Jones's theme from The Best of Both Worlds a lot better, anyway...)
[Spoiler below, hopefully not a huge deal for an almost forty year old movie)
Voyager returns under the name v-ger.
I have a vague recollection of the humanoid that V'ger hijacked to serve as a voice being romantically involved with the original captain that Kirk replaced, and that relationship being exploited and becoming a plot element later. So I think there was some going on...
Sad thing is I saw the movie in the theater when it was first released...just didn't put the two together.
The NT machine was called Locutus.
Related question ... can an easily amateur listen to response transmissions like this that come back from probes?
Has anything like this been tried before with other spacecrafts?
You'd need to totally own the DSN. According to the Wiki:
Because of the enormous distances and the resultant weak signals from the spacecraft, the large antennas and the very sensitive receivers of the DSN are required to provide the necessary communications capabilities. The DSN is the world's largest and most sensitive spacecraft communications network. It consists of three deep space communications complexes located approximately 120 degrees of longitude apart around the world: at Goldstone, California; near Madrid, Spain; and near Canberra, Australia. This placement permits continuous communication with a spacecraft.
Perhaps listening to the traffic could help deciphering the encryption (if any).
What it wouldn't be easy at all is to have antennae large enough and in various parts of the world: amateurs still don't have the capacity to do that.
> Has anything like this been tried before with other spacecrafts?
The only thing I can remember is reviving an old satellite with linux and SDR (https://www.ettus.com/blog/2014/07/contacting-a-36-year-old-...) but that was with some cooperation on part of NASAs.
“The best known of alleged takeovers of satellite control occurred in 2007 and 2008. In particular, a serious attack was observed in 2008 when hackers obtained the control of the NASA Terra EOS earth observation system satellite for 2 minutes in June and for another 9 minutes in October. Fortunately the attackers didn’t damage the satellite during the time they gained control of it.”
And then the 20kW+ transmitter you'd need...
There were absolutely compromises, and the engineers who shipped it knew myriad ways in which it could be better.
To answer the question:
Find something you think is equally cool, learn what you can about the field, be open to the possibility of a pay cut, and start knocking on doors. Sometimes doors open. Work really hard when they do.
Although it could be attached to a weather balloon and still be useful for scientific testing.
Ok, the pay isn't silicon valley level but most labs take work life balance and professional development quite seriously and you get to work on cool stuff.
One thing I would caution (from my experience) is that the attitude was very much "get it to work" vs "do it properly". I was writing software and had quite a bit of freedom, but the hardware guys were often tasked with, "We have this problem and there is nothing that we can buy/afford that will solve it. Could you please invent something? You have that mountain of spare parts from previous projects to work with". They would be wandering around the department saying things like, "Do you need that <whatever piece of electronics> any more?", and then would steadfastly break it down to get the parts/components that they needed. Probably less reusable stuff in electronics these days, but I remember being pretty awestruck at the time by what they could scavenge and then build.
I remember in the physics lab, my supervisor/boss had bought a laser printer from Europe. Because it was 240 volt, they built a voltage doubler so that they could use it. One day they accidentally plugged it into 2 voltage doublers and... well, predictably it no longer worked. But my buddy quite happily pounced on the dead carcass and I don't think it lasted more than a couple of hours before the working parts were repurposed for something else.
Like I said, this was a long time ago, but I imagine things still mostly work the same. Awesome place to work.
The tension between "get it to work" and "do it properly" is still there, particularly when there is pressure for results for conference papers. We try to do things properly though, simply due to the more onerous safety regulations now.
It is interesting coming across various historical bodges that lurk in dark corners, sort of little engineering time capsules.
Of course, you can work on whatever you want to whatever standard you want if you do it on your own time & dime, if you have the willpower and the budget, or can find a patron, crowdfunding is viable now.
My favorite use of tax dollars ever.
-- Many materials age on their own; plasticizers outgas from plastics. You're probably familiar with how once-pliable but now old/aged plastics, even kept sealed in a box, can become crunchy and brittle.
-- Chemistry doesn't stop in space. The thrusters are likely to involve chemically-reactive materials. Any little bit of corrosion, stress-corrosion cracking, etcetera could cause anomalous performance.
It is wonderful when things still work.
See e.g. this study on plastics erosion in LEO: https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/201700...
Not sure how it varies as you leave the solar system - depends on the exact balance between Solar Energetic Particles (particle radiation from the sun) and Cosmic Rays (particle radiation from outside the solar system), and on how much the cosmic radiation increases as Voyager leaves the heliosphere.
Also, at such an enormous distance, I’d expect very minor dish positioning errors to result in the loss of the line. It’s awesome that they had the skills to build something like that in the 80s.
The idea for it (a "grand tour" of the planets due to a particular orbital alignment that made such a thing ideal) was hatched in the 1960s...
So this is the same problem as SETI. I know for SETI they reduce stellar noise by assuming that aliens are transmitting in very narrow bandwidths.
It is written by lisper at HN who submitted this article :)
It's not pointed in the right direction, but it's interesting to think of this as just over 1/2000th the distance to the nearest star.
It'd be cool if right before we lose contact of it for good, we got it to go as fast as possible so that it will reach who knows where someday... just slightly faster.
The additional speed we might get out of it is tiny compared to the 17km/s it already has.
If they used what little fuel was left to go faster, then wouldn't be able to communicate with the Earth. The current goal is to get information about what's at the edge of the solar system.
We know it's not going anywhere close to anything, any time soon. Wikipedia says "in about 40,000 years, it will pass within 1.6 light-years of the star Gliese 445".
And for all we know, if it goes faster it might pass by some destination it would reach now.
It's in interstellar space. Barring a really lucky encounter with another interstellar body, it'll be another 40,000 years before it's close to anything.
We are finding more trans-Neptunian objects. There's about 2,500 of them. But as Adams said, "Space is big. Really big. You just won't believe how vastly, hugely, mind-bogglingly big it is."
Voyager can only be angled a smidgen from it's current path. There's nothing it can reach to do a slingshot.
If Voyager could change its path by 0.1 degree (which it can't), and if all of the trans-Neptunian objects were equally distributed around the Sun (which they aren't - and Voyager is going out of the plane of the ecliptic), then that's still only a 0.2% chance of having something in its path.
Even if there were, Pluto gave New Horizons about 5-6 m/s boost. That's effectively nothing compared to Voyager's 17 km/s speed. https://www.youtube.com/watch?v=Hm6ga-g9ACU via https://space.stackexchange.com/questions/10087/did-the-plut... .
anyone have more info on the power with which the signal arrives here at earth ? and how do they make sure that ambient/thermal noise does corrupt it ? thank you !
As to why there's propellant left: it's a sealed system and NASA has intelligent engineering practices when it comes to resource planning.
Still, a stupendous achievement to have them still working 40 years after being deep frozen in space!
> Combustion Chamber Pressure: 4.35-23.9 atm. Combustion Chamber Temperature: 800 Celsius. Combustion Chamber Cooling Method: radiative. Duty Cycle: 0.008 sec minimum to unlimited maximum burn. 750,000 pulses.
So they probably didn't bother calculating how much fuel to put on board for the planned trajectory. They used the same thrusters for New Horizons!