Hacker News new | comments | show | ask | jobs | submit login
After 37 years, Voyager has fired up its trajectory thrusters (arstechnica.com)
740 points by lisper 8 months ago | hide | past | web | favorite | 260 comments

Voyager is probably the most badass awesome thing ever.

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.

What I find more amazing is the fact that a lot of what could be referred to as "space-age technology" is actually many decades old, and thus was accomplished with a fraction of the processing power and knowledge we have today. Voyager was launched in the late 70s, but based on technology of the 50s and 60s. We visited the Moon almost 50 years ago, using that technology.

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.

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.

>I’m currently travelling at about 180mph on board a high-speed train in Japan.

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.

I agree that LTE/HD is incremental. It's probably fair to say that ubiquitous mobile smartphones are something more. (Initial mobile communications less so. When I had a feature phone I often didn't even carry it with me although certainly some people probably found them more transformational.)

You may have missed the last word of his previous sentence. reliability. Which is at least to me, the way I understood it.

Something built back then and still works.

Edit: fix markdown syntax.

But they're different systems, designed very differently. A high speed train and a space craft are both designed carefully to maximize their possible life. A cellphone is, unfortunately, been phased into the economy of consumption and planned obsolescence.

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.

Fair enough. I haven't had my morning coffee yet, but I do agree. You make a good point. Thanks for correcting me / explaining.

Will deffo check the book out.

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.

Marry such a nuclear reactor to something like VASIMR engine, and suddenly the entire Solar System opens-up to us.

The reliability of these systems is indeed quite impressive, yet it simply isn't a requirement for most of our day to day equipment. OTOH some heavy duty machinery can be very reliable.

I'm pretty sure today's spacecraft are very reliable too.

We always building on what was previously there so one can make that argument for almost about anything.

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.

Focused human attention is still by far the most powerful optimization device we have.

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

I've lived through a couple of design reviews for flight software and hardware, and I can tell you this is spot on. The emphasis is on minimal delta from previous products, absolutely nothing remotely risky, and complete assurance of functionality before launch.

Have you heard of the Opportunity Rover? I believe it has outlasted its designed lifetime a few times now. Also, its still operational.

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.


>What I find more amazing is the fact that a lot of what could be referred to as "space-age technology" is actually many decades old, and thus was accomplished with a fraction of the processing power and knowledge we have today.

Yes, but then they weren't using it to power immense social networks with pictures of people's food.

They say these new thrusters will extend Voyager's lifespan by 2-3 years.

Just curious but does anyone know it's mission in interstellar space? Is it more than just a fun experiment at this point?

Both Voyagers are still collecting magnetic/plasma field and radiation data (only Voyager 1 is in interstellar space, Voyager 2 hasn't made it yet). Mostly they are measuring the outer edges of the Sun's magnetosphere as it interacts with interstellar space, and it's a dynamic phenomenon so the data is actually more interesting than you might imagine. Several research papers a year continue to come out of the missions.

Radiolab did a great summary of this just a couple of months ago: http://www.radiolab.org/story/sun-dont-shine/ You can skip the parts about the time capsule if you want.

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.


To explore strange new worlds, to seek out new life and new civilizations, to boldly go where no machine has gone before.

V'Ger will come back.

aren't they on hyperbolic trajectories?

Well spoken!

For anyone with an interest in Voyager, I highly recommend Emer Reynolds film, "The Farthest". This documentary is brilliant, beautiful, and funny - and I guarantee that you will learn something that you didn't know about Voyager!

For those in the UK, it's currently streaming on iPlayer - though it should ideally be seen in cinemas to be fully appreciated.


In the US, the film is on Netflix. I haven’t seen it yet so I can’t recommend it personally, but I’m looking forward to watching it.

Gotta love Netflix. They have a pretty solid science documentary collection

I saw it when it aired on PBS. I agree with GP, it’s fantastic.

Torrent link for where it's not available: https://1337x.to/torrent/2379692/The-Farthest-2017-WEB-DL-x2...

.onion link to torrent:


Magnet link:


Thank you for this recommendation - just finished watching this on iPlayer and it was everything you promised.

I came to say thank you. I just watched it on netflix, it's brilliant, and funny.

So, they have all the codes and manuals to be able to control the probe after 37 years. More than that people able to operate those manuals.


From https://www.jpl.nasa.gov/news/news.php?release=2017-310

> "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'm curious as to what they mean by 'outdated.' Is the actual language outdated or simply the architecture? I would assume the latter, but it's tough to tell. I just can't figure out how assembly could become outdated.

Each processor has its own instruction set and thus assembly. Further, there are various assembly syntax dialects that can fall in and out of use. For example, the AT&T syntax and Intel syntax used for x86: https://en.m.wikipedia.org/wiki/X86_assembly_language#Syntax

Meanwhile, after two years my Android phone will brick itself.

> Meanwhile, after two years my Android phone will brick itself.

I understand the sentiment, but your phone was not made to exit the solar system.

But they should be .. or at least last as long as an old C64. You can put newer Windows on pretty old hardware, and for really old hardware you can always slap Linux on and it will still be useful. There are still Kernel forks to support 386 processors!

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.

Agreed, but if a public company makes stuff to last 10 (why not 20) years then they'll go bust because capitalism requires profit and sustainable ideals are contrary to profit.

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.

It's not the lack of a platform, SBCs share operating systems and updates because there are mechanisms to deal with this.

The problem is the lack of vendor participation in the community.

<Insert Samsung Galaxy joke here>

Samsung Galaxy was designed to simulate the end of the solar system.

If it's anything like my old Chevy Nova, it'll light up the night sky.

I would think it was meant to simulate a super nova.

Note 3 with 3 gig ram. Works great.

Neither was Voayger

That's a very good point actually. It was meant to flyby the outer planets, and its mission was extended when it refused to break down.

Officially you're correct. But from the start the people working on it had planned/designed it to go further.

Opportunity was designed to work for about 3 months. It's been going more than 13 years.

Instead, was designed to last only as long until the next induced appetite for a new phone. Cuz otherwise, financials wouldn't look as good while another company X in the industry capitalizes on it anyways

It was cheaper too.

It took my Nexus 40 days

A little difference in design specifications and requirements :)

And cost.

Indeed, Voyager 1 and 2 combined cost slightly more than half of an iPhone X.

With orders of magnitude less processing power. Government projects just cost more..

Let's not talk about iOS updates that is making 6, 6s and 6s plus deprecated by lag

This is a great article on the engineers who have spent their careers caring for the Voyagers


Thanks for the link, a very nice article indeed.

NASA usually has a duplicate system (full mechanicals, not just a simulation) to test things on to make sure they don't create a brick. NASA can also use the dupes as a backup in case there is an accident during launch and the first system gets destroyed.

This assertion is correct for the more recent robotic missions (Mars Science Lander, Mars Exploration Rovers (MER), Mars Pathfinder/Sojouner) but I don’t recall seeing or hearing about spares for earlier missions or satellite/spacecraft-based missions. Earlier missions would probably have had only an empty mechanical flight spare. I worked in the Spacecraft Assembly Facility at JPL and was very familiar with the Technical Inventory Management team (the group that managed storage and logistics for reusable and non-reusable technology assets).

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.

I wonder how well the 37 year old duplicate mechanicals work.

Well... as long as they work precisely as well (or badly) as the 37-year-old originals do, that's perfect, right?

They should be kept in a space-cold vacuum.

The point really isn't to make sure the valves work (the environment might have broken them after a few decades in space), the point is to make sure the code you're uploading doesn't brick the computer. It's nice to have an original computer and fully replicated signal path to make sure there aren't any unforseen bugs. Emulators have a tendency to have 99% coverage but the missing 1% is what would cause us to throw an expensive brick out of the solar system.

they have the source too I suppose? they probably created a simulator to verify that they understood everything well, and because 19h is a long time

It's a lot easier when it's literally impossible to change the functionality.

The voyagers can be upgraded remotely. It had too, as it doesn’t have enough memory to store a program for its full mission.

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)

> It had too, as it doesn’t have enough memory to store a program for its full mission.

Don't you just hate it when they make devices rely on cloud services... :P

Not true, the voyagers can be in-flight reprogrammed: https://history.nasa.gov/computers/Ch6-2.html

They can not only be remotely upgraded, but remotely reprogrammed to avoid bad bits. 1970s computers were primitive by our standards, but not that primitive.

In the Epilogue of "Murmurs of Earth" (1978), Sagan writes:

"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.

I could easily be wrong here, from the article it sounds like the remaining power reserves are only enough to correctly orientate Voyager to allow for its communications link to be pointing towards Earth.

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 :)

60,000 years ago, humans were just leaving Africa. So I would be impressed if 60,000 years from now, anyone remembers that Voyager is still out there!

I really wish I could see the progress humans make over the next 60,000 years. Shout out to all my ancestors in 62017!

I think you mean descendants.

There was an accident involving a contraceptive and a time machine.

Very likely there won't be humans in the next 100 to 500 years, assuming any reasonable rate of progress. Humans will quickly change themselves biologically, technologically, we are going transhuman.

That's very optimistic. There are are two pretty likely outcomes: humans will solve our problems, bring about peace, colonize Mars and voyage out into space ... or... we go extinct. We might waver between tech and stoneage for a bit with some war, but eventually we're likely to converge on one of those two.

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.

At the end were just apes with super computers. 100-500 years is not a long time.

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.

Sure we will, in my opinion, Forgotten is the biggest weakness we human have.

Just watching those kids quantum stream their lifes in the Andromeda, yet none of them even mentioned the good old Voyager 1.

You are standing on the shoulders of a giant star with your writing there.

Kudos to Aerojet Rocketdyne on their thrusters lighting up after 37 years. That's genuinely impressive engineering.

Certain Volkswagen models apparently have an even more amazing MTBF. It's a real sleeper. [1]

[1] https://www.youtube.com/watch?v=Xo2kSu6O8cU

But Voyager's got it beat on mileage...

Not just 37 years, but 37 years in near absolute temperature.

Can somebody with knowledge of radio communication explain how we are able to send a radio signal to a destination that is 21 billion Kms away? How powerful does the signal need to be? What kind of technology is used to generate such a powerful signal?

It's mostly a function of how well your antenna is able to amplify the signal, how precise you can aim that antenna and how slow you transmit. If you have a very good antenna with lots of gain and a very precise mechanism of aiming it (at an object whose location is very well known, and Voyager is a moving target) it will take much less power than if either one of those elements is not optimal.

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.

If we launched a proxy satellite into the solar system to help communicate with these probes, would it make things easier/better? Or is it better to just communicate directly from the Earth?

For Voyager-like mission profile building useful network of such data relay satellites/probes would be prohibitively expensive.

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.

Isn't Voyager on an an angle relative to the ecliptic that is large enough that a single satellite could have it in view at all times?

See the other thread :). Such satellite would need impractically large Rx sensitivity and Tx power. Doing that on surface of earth is simpler and significantly cheaper.

Are there bots that downvote comments on HN? Why would anyone downvote this comment?

I think it would make things worse, the only advantage I see is that if you got it to work at all you could do away with the requirement that you'd need three base stations on Earth. A single Geostationary satellite should be able to keep both the base station and the Voyager in sight at the same time. But that's a lot of money to spend for very little advantage.

There is no launch vehicle that is capable of launchibg anything remotely similarly sized as typical DSN antenna, not to mention of similar mass. Also station and attitude keeping of such an satellite would be non-trivial problem at sufficiently high orbits for such thing to be useful.

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

It would not nearly have to be as heavy as it would have to be on Earth (no gravity to withstand so skeletal build will do just fine) and a 20KW transmitter could be powered on a satellite with relative ease. There have been quite a few proposals for antennae that unfold in space.

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).

Unfolding some lightweight structure is certainly possible, but the attitude-keeping requirements are harsh.

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.

> but the attitude-keeping requirements are harsh.

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.

My point was that the actual payload is mainteance-heavy.

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.

Interesting! Thank you for the info, I'll ruminate over that for a while!

There are high gain antennas on both ends, and use of error correcting codes, as well as extremely high grade low noise amplifiers on the Earth side.

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

There's also a neat book for those interested in the engineering details of their receivers:

"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.

Cool video, thanks! They also mention this website[1] where you can see the status of the whole DSN live, what antennas in what locations are communicating with what spacecraft, with signal strengths and transmit powers and bitrates and all.

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.

[1] https://eyes.nasa.gov/dsn/dsn.html

That's amazing.

It's obviously a question of power and antenna gains, but a very important role is played by the encoding: Pioneer before, and Voyager after introduced concatenated codes [0], which consist of a viterbi decoded convolutional code inside an outer Reed Solomon code.

Curiously, those codes are now superseded by other, more modern approaches, for example Turbo Codes [1] which are used not only in deep space probes, but also in cellular communications and other applications that we consider normal these days.

[0] https://en.m.wikipedia.org/wiki/Concatenated_error_correctio...

[1] https://en.m.wikipedia.org/wiki/Turbo_code

I worked as an RF engineer for a few years and one of the senior engineers once told me a story about how they launched Voyager (I think it was Voyager) with a Viterbi encoder, even though we didn't have the ability to make a decoder at that time due to lacking the computing power. So they launched it knowing they would eventually have the technology, and of course they were right. No idea if that's true or not but I love the story.

I read those links because I'm really curious but I clearly lack the prerequisite knowledge.

According to the inverse-square law, doubling the distance will reduce the power of the signal by 1/4th. The moon is about one light second away, whereas Voyager is about 70,000 light seconds away. If we round down to 64k light seconds, that's about 16 doublings. If each doubling represends a 6db loss, that's about 96 decibels.

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.

That's assuming you're firing the signal out in all directions though, right? For example, a laser doesn't drop in power in this way. I would think a tightly focused radio signal could also avoid such extreme degradation.

It's really hard to make electromagnetic radiation perfectly parallel. If you shine a laser pointer at something far away, the dot gets bigger with 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.

> According to the inverse-square law

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.

https://space.stackexchange.com/questions/9824/how-much-rf-t... A good bit but not a ton seems to be the answer. Maybe in the range of a radio station.

The more interesting part is that Voyager can send signals we can receive (with giant, huge radio dishes, but still).

I'm not sure that the more interesting bit is the voyager->earth part.

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.

Our end is not stuck with 40+ years old, non-upgradable electronics. The Voyager end is where the magic is.

Exactly my point. The fact that voyager can still hear us, with 0 upgrades, is far more impressive than the fact that we can still hear voyager.

> 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).

and Voyager is listening using 40-50-year-old technology, its far more impressive that voyager can still hear us than that we can still hear voyager.

We can yell louder, listen harder, and pour more computer power into decoding the signal than Voyager can, though.

It’s mostly just very high gain antennas in both directions. You don’t need all that much power if your data rate is very slow and you have a very narrow beam.

Does anyone know what is huge baud rate for voyager? Dial up?

In the neighborhood of 160 bits per second I think for telemetry. There is other information.


At a very low baud rate.

He said that ('slow data rate').

I was wondering the same thing and found this:


(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.

Remarkable. I'm vividly reminded of that scene in the movie, Apollo 13, where this is said: [Gene Kranz:] I want you guys to find every engineer who designed every switch, every circuit, every transistor and every light bulb that's up there. Then I want you to talk to the guy in the assembly line who actually built the thing. Find out how to squeeze every amp out of both of these goddamn machines.

NASA has a remarkble group of engineers who know how to get every last erg of energy out of that machine.

I think Apollo 13 did a lot to show what talented engineering teams can do. They also didn't shy away from real terminology and rarely stopped to explain it. No other movie has done that in my experience.

Very impressive. Every day still the envelope of what mankind has touched is still expanding because of this craft. It's a tiny mark on a vast universe but for some unspecified reason it makes me feel very happy to know that it is out there and still ticking, I am not looking forward to the day that it eventually will shut down but even as an inert man-made mass that far out it will be an amazing accomplishment.

I am not looking forward to the day when it somehow returns, starts blowing up star ships, ultimately taking over one of them, and tries to seduce the crew

Or start cheering for nationspanning American football games with other santient probes.

17776 Football was the best thing I've experienced in years. I'm not even into sports. Tingles just thinking about it.


That... that was really something. That was and is incredibly engaging for me, though it's quite different and I'd understand if someone didn't find it to their taste. What a totally beautiful artifact of the internet.

I discovered this thanks to you. I'm still reading it, but I just wanted to thank you. This is brilliant stuff.

What....what is this?

New plan for the evening...

This is cool. Thanks.

I can see a Gravity Gate happening when Andromeda Patriots tweak the Earth gravity simulator of their home stadium to gain an advantage.

That's actually a great movie plot, not sure if it has been done before.

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

Here's another one for you:

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

Upvoted purely because I love the fact that there's someone here who hasn't seen ST:TMP.

(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...)

That is basically the exact plot of the very first Star Trek movie

Spoilers below:

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.

More spoilers below:

> alien machine intelligence

The Borg.

That's fanon. Yeah, I sort of like the idea too, and yeah, Goldsmith to some extent may have styled the First Contact Borg leitmotif after V'ger's from TMP, and yeah, maybe he did that deliberately, and yeah, somebody's friend of a friend said that Gene Roddenberry once coughed in a way that could be interpreted as suggesting he'd had in mind the Borg creating V'ger all along - but it's fanon all the same.

(Me, I like Ron Jones's theme from The Best of Both Worlds a lot better, anyway...)

The connection to the Borg isn't cannonical. Though Star Trek's cannon has gotten extremely messy in recent years.


Lucky then that the second Voyager gets blown up by a wildly overacting humanoid alien with mad hair

More like Singer flings a Dual-Vector Foil at us...

That's actually the plot of the first Star Trek movie (without the seduction part).

[Spoiler below, hopefully not a huge deal for an almost forty year old movie)

Voyager returns under the name v-ger.

> without the seduction part

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...

I think that was the joke that GP was alluding to :)

Back in the early 90's I wondered why so many Unix boxes had 'vger' in their name. Not until this year, when I watched the original Star Trek with my youngest, did I realize what it was all about.

Sad thing is I saw the movie in the theater when it was first released...just didn't put the two together.

The first two Linux machines I ever had to admin back in the 90's where called Picard and Kirk (Picard was the primary, Kirk the secondary (no guessing which series I preferred)).

The NT machine was called Locutus.

The "Locutus" is so perfect!

The art of educating people is to helpfully answer questions as they arise, not to interrupt the good mood with a display of your knowledge.

Pretty sure that's the joke.


It's funny that when considering the current dire and dispiriting geopolitical situation, it's stories like this that bolster my faith in mankind.

The round-trip request/response time for a command is 39 hours (!) ... I guess the team has normalized to that latency, but when asking a system to fire with paths of execution plus hardware plus propellant/etc that haven't been used for 37 years ... I can only imagine the tension and celebration.

Related question ... can an easily amateur listen to response transmissions like this that come back from probes?

The Deep Space Network uses a 34-meter antenna to contact Voyager, not something most people can put in their backyard :) https://science.howstuffworks.com/question431.htm

I had a chance to visit the Deep Space Network facility at Goldstone several years ago (definitely worth a visit if you can arrange one). The antennas they use vary in size from 26 to 70 meters in diameter, and even then they need to use lots of process gain because the signal is still several dB below the noise floor. Amateur tracking would probably be incredibly difficult.

You will definitely want to be doing cryogenic cooling of your receiver front-end to reduce the noise figure[0].

[0] https://en.wikipedia.org/wiki/Noise_figure

Just as a reference point, GPS signals are below the noise floor as well. Typically by 20dB.

Googling Amateur-DSN will probably give you a link to the Yahoo group where the people who do that kind of thing hang out. The short answer is that yes, numerous interplanetary probes can be received by amateurs with moderate effort/expenditure, but reception from the Voyagers is out of the question without access to a massive antenna.

In the spirit of what this community's name suggests it's about, how hard would it be for someone to hack into the spacecraft by sending it a signal to turn on its thruster too soon given the technology is 40 years old?

Has anything like this been tried before with other spacecrafts?

"how hard would it be for someone to hack into the spacecraft by sending it a signal"

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.

I guess along with knowing the signal protocol you would need to know the exact location. Security via obscurity?

> In the spirit of what this community's name suggests it's about, how hard would it be for someone to hack into the spacecraft by sending it a signal to turn on its thruster too soon given the technology is 40 years old?

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 NASA Terra Earth orbiter was successfully attacked in 2007/8 — http://resources.infosecinstitute.com/hacking-satellite-look...

“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.”

Know the big dish from James Bond? Used legit: http://observer.com/2014/08/civilians-in-abandoned-mcdonalds...

Nope, Arecibo is not used for this. The Chinese FAST would be perfect for this also.

You are correct this wasn't the article I was looking for, didn't read it to verify. I can't find it now but 1-2 years ago some people used that dish to reactivate a "lost" satellite that had not been used for several years.

I imagine that with cracking the protocol with 39 hours return time... it would be a challenge.

At a mammoth 16bps...

And then the 20kW+ transmitter you'd need...

I think Voyager data rate is 160bps (when I last saw it at https://eyes.nasa.gov/dsn/dsn.html ).

It wouldn't surprise me if they've been reducing the data rate as the signal strength weakens.

Yep. But looks like Madrid is receiving data right now from VGR1 and the page says "159.00 b/sec".

Wild engineer's dream: where can I work on something of the same coolness level? High scientific content, total quality, no compromises whatsoever.

I'll burst the quality bubble now:

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.

Two months before the Voyager probes were shipped for launch it was found out that Jupiter's magnetic fields and radiation environment were a lot stronger than anticipated, in excess of what the spacecraft were designed to tolerate. They ended up wrapping many of the cables on the spacecraft with aluminum foil bought from a local supermarket as a protective measure.

Also engineers love constraints and compromises. It's a nice dance between getting shit done and getting shit done right.

Does it need to be a job? If not, you can build your own low orbit satellites, scale rockets, satellites communications, aquatic robotics and more. In fact, aquatic robotics are one of the most approachable ones because you can do it in a nearby lake/river/beach and do proper science. Plus you can team up with local scientists and help them gather/process data.

You just gave me so many ideas for fun side projects.

Feel free to ping me for any more ideas or help :)

can you really build a satellite as a hobby project ?

Yes, there are things like cube-sats. Problem is affording the launch price. ;)

Even building a cube-sat that never leaves your desk would be an interesting hobby project, really. Non-experts would certainly learn a ton from doing so, and you'd end up with a cool conversation piece, if nothing else.

Absolutely. It's why building one is on my todo list.

Although it could be attached to a weather balloon and still be useful for scientific testing.

I've heard they really need people who'd be willing to replace the aging programmers who have been overseeing Voyager for a long, long time...


You don't say what kind of engineering you do, but you could do worse than working at a particle physics lab. You have everything from data analysis, complex control systems, physics simulations, high voltage electrical engineering... the list goes on.

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.

I just want to say that I've literally done this and it was awesome. Granted it was when I was at university and it was a "part time" job, but I've often thought about going back. Now this is going back 30 years and academia in NA has probably changed a lot, but at least at that time most of the top research groups had enough money to hire full time "technicians". At my university it was a union position, so the pay and benefits were not bad compared to the rest of society. Of course compared to the high tech industry, the pay was not even close :-) Probably for that reason, every body I worked with was passionate, brilliant and more than a bit quirky. Anyway, I ended up working in a chemistry group doing crystallography, a physics group doing high energy physics and an astronomy group that was working on community education programs. Each team had really interesting challenges.

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.

I currently work for a UK based particle physics facility and a lot of what you say is still true, it is an 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.

aerospace (airliners, hardened electronics for space, etc) / robotics (all those problems with the fukushima robots, precision medical and industrial bots) / medical devices of all kinds

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.

Probably nasa or spacex?

I love that every time I read one of these articles it is extending the life of the mission by 2-3 years. :)

My favorite use of tax dollars ever.

What do you mean? Any context here?

When you're far enough from the sun to avoid thermal cycles, in the vacuum of space, what forces will cause things to deteriorate? If I wanted something to last forever, deep space would be the perfect place for it.

-- It is a tough radiation environment. Materials are embrittled and altered over time.

-- 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.

And then there are vacuum welding and interstellar dust.

Interesting, I didn’t realize cold welding was a problem in space. I assume it can be solved by applying a coating to surfaces that might come in contact with each other.

Radiation, mostly - though close to planets you also get some chemical erosion from high-speed gas molecules.

See e.g. this study on plastics erosion in LEO: https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/201700...

Does the radiation decrease as you get farther from the sun? What's the background radiation in deep space vs. here on Earth?

It definitely gets higher when you leave Low Earth Orbit, because you're leaving the Earth's magnetic field; so the ISS is definitely at the low end of the scale of Voyager's exposure.

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.

I believe the thing that will kill voyager is primarily the radioactive decay of it's battery. It's power budget is so low now that it's got most everything off almost all the time. Sadly there is a point where they won't be able to turn on the radio to listen.

Isn't the more immediate concern that the power supplies are going to run out?

According to Wikipedia the atomic batteries are good until 2025, which isn't so far away anymore.


How does Voyager know where Earth is, such that it can position the dish correctly? Does it use inertia to keep track?

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.

It doesn't need to point its dish at Earth, it can simply aim for the Sun. It is so far away they are as good as in the same position relative to itself. And the Sun conveniently lights up making aiming a lot easier.

Looking up the math on that, the high gain antenna is within 3dB of max at up to .3 degrees off-target. It's 132 AU out, so Earth is within .41 degrees of the sun. Good enough!

Did they have digital light detectors in the 70s? I mean, it’s trivial now to aim something at the sun now, but they didn’t have image recognition then. A primitive heat sensor perhaps?

Yes, CDS cells are much older than that, and Ge diodes are light sensitive too.

Voyager 1 was launched on September 5, 1977 (notably after Voyager 2 was launched!).

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...

Even more impressive, Voyageur 1 was launched in September of 1977 [1]. It was originally built on early 1970's technology. An incredible amount of foresight and luck.

[1] https://voyager.jpl.nasa.gov/

Voyager's antenna is aimed at the sun (because at its distance, earth and sun have close angular separation). So this means the command signal has to overcome microwave noise from the sun.

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.

Fascinating. Now, that's what I call Engineering. And +1 to the author for using the metric system :)

Lisping at JPL (Jet Propulsion Lab) is an interesting read: http://www.flownet.com/gat/jpl-lisp.html

It is written by lisper at HN who submitted this article :)

The speed of light sucks. What patience.

19 hours and 35 minutes * speed of light is a long way.

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.

Would it be possible to use these thrusters, or even the main thruster, to speed Voyager or slingshot it around something right before it loses power to communicate back to us?

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.

As ssijak pointed out, there's nothing to slingshot around.

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 has long past sling shot phases around planets

> slingshot it around something

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.

"Close" here meaning it will be 700 times farther away from that something than it is from the sun now.

That puts it in perspective. Space really is huge.

I mean, we are finding Pluto sized objects right outside our solar system all the time now.

It doesn't seem like it's all the time. Pluto is the largest, Eris is close. Haumea, (225088) 2007 OR10 and Makemake are the only others at least 1/2 the diameter of Pluto.

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... .

You forget how big space is.

just as a scale, this distance is not even a light-day away ! i am just amazed that we can still "hear" it.

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 !

You can normally see them communicating here: https://eyes.nasa.gov/dsn/dsn.html

There's a wealth of fascinating information including book series on deep space communication at the JPL DESCANSO Deep Space Communications and Navigation Center of Excellence:


What does it fire out when firing the thrusters? How come there's propellant left?

The thrusters use hydrazine. It reacts with a catalyst and decomposes into hydrogen and nitrogen. They are used to steer, so the fuel isn't used up.


Well, cough, if rocketry is involved then reaction mass is being expended somewhere somehow... whatever the underlying chemistry may be... so it is a finite resource that is subject to depletion...

Oh relax. They are used to steer and not used frequently, so the fuel hasn't been used up.


I think you're saying 'the fuel hasn't been used up' to mean that 'the fuel has not been entirely spent and there is plenty left', and the other person thinks you mean 'no fuel is used at all when the thrusters are fired' which is why they're correcting you. Both are reasonable but different interpretations of 'used up'.

Yep, that's what the statement looked like it meant to me on the first reading.

A clue to my meaning was that I also mentioned a catalytic reaction.

I'm an applied mathematician and a practicing economist, I'm afraid I wouldn't understand chemical reactions as being of one sort or another if they hit me on the head and set me on fire. I'm honest, I know my limits, and chemical ignorance is one of them.

Much better, yes.

They are pressure fed hydrazine monopropellant thrusters. They are incredibly reliable. Keep in mind that Voyager 1 doesn't have reaction wheels or any other method for controlling its attitude other than its thrusters, and it's 3-axis stabilized. The original thruster set fired thousands of times over a nearly 4 decade period and now the backup thruster set is being rotated into use.

All the references I found Wikipedia said it used Hydrazine thrusters.

As to why there's propellant left: it's a sealed system and NASA has intelligent engineering practices when it comes to resource planning.

How could a rocket system, which when in operation vents reaction mass into the vacuum of space, ever qualify as a “sealed system”? Surely only the storage tanks could ever qualify as being a “sealed system”, and even then an imperfect one, because there must be one or more potentially leaky valves leading to the actual nozzle somehow?

No seal is perfect, especially without standard pressure surrounding it.

The fact that there is propellant left implies that the seals were good enough

If you wait for the perfect seal you will never launch. And it will cost you a lot of fish.

I always thought "thrusters" were devices that always spewed out majestic flames and plumes of smoke whenever they are fired. I was surprised, as a kid, to learn that the Voyager thrusters effectively are no more exciting that pressing the nozzle of an aerosol spray can for a split second. Stands to reason that they don't need much force to affect the spacecraft inertia at those weights and speeds.

Still, a stupendous achievement to have them still working 40 years after being deep frozen in space!

If you read the specs (http://www.astronautix.com/m/mr-103.html) of the thrusters in question, you can see that they do "spew flames" and there is a combustion chamber:

> 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.

But from what I remember reading in the past, those flames are essentially invisible in a vacuum, is that right?

Irrespective of what is used, consider that in the vacuum of space, when you're not in a hurry, you'd only need the tiniest bit of propellant/thrust for an attitude correction. Even for a comparatively heavy vehicle like voyager. The other forces at play are so miniscule, that you only need a ridiculously small force to start it rotating (and the opposite when you want it to stop).

propellant, and because they didn't use it all.


Interesting that it will run out of electricity before it will run out of Hydrazine.

3 ounces of thrust each:


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!



Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact