I feel this is one of those moments when the technical term "whew" is appropriate.
I was born in the early 70's and the Voyager's have been travelling most of my life. In some ways they're like a touch-stone, something I always click on when I see an article.
One day, like me, they'll finally die, but until then may they travel onwards and outwards.
A tip of the hat to folks at NASA and JPL that built this marvel, with 70's tech, that still continues to amaze and delight us.
I work on spacecraft flight software and I was never worried about this lol
The worst that could have happened is that the spacecraft would have kept the dish in the wrong position until October, at which point it would have re established contact. This isn’t even the last time the Voyager program has had similar outages, IIRC they had to shut down the dish for maintenance for several months and when they re established contact the spacecraft was fine. They should review their MOPS procedures so this doesn’t happen again (it’s generally standard practice to run commands against a simulated version of the spacecraft in a similar operating condition or a replica to make sure it won’t do something destructive) but from the first time I saw the headline for this my first thought was “lmao” and I was never really worried this would cause lasting problems for the spacecraft.
Things worked out this time, but the cavalier attitude is dangerous. I also work on spacecraft flight software and this kind of contingency is the one you should exercise only in sims. Voyagers are unique hardware that you can't ever service - the correct risk posture here is don't.
Some of the risks involved are - we don't have a good model of the flying hardware. The telemetry is extremely limited and engineers who made / understand it are scarce. Current situation of messing up commands is one of the consequences of that. There is a number of faults that are isolated and recovered via software workarounds, any kind of reset could default you in a non functional configuration. This is aging hardware that is slowly deteriorating, severed comms stop your ability to react to the events.
The overall risk of permanent contact loss was probably low, but due to all the unknowns the error bars are quite high.
Im already always much afraid when I have to remotely reboot a workstation and delighted being able to ssh into it again soon later, though the worst case here is not as bad as with this most far away remote machine, thus amazed by your confidence and calmness ;)
The software is much simpler on Voyager so there’s not a lot to go wrong and there are a shitload of failsafes that have to go wrong before the vehicle stops working. For example on some stuff I’ve worked on and the Mars rovers there are multiple boot chips per flight computer and multiple flight computers per vehicle, and even if both are dead you can do some basic tasks just using a micro controller on the radio.
One interesting read from a purely software side is the JPL C standards, which are an extension of software standards for things like Mars rovers that cannot not work. There’s a summary on page 4 if you don’t want to read the whole thing https://yurichev.com/mirrors/C/JPL_Coding_Standard_C.pdf
But unlike you and I, when they die they'll continue as a fascinating relic of our species and our time. And in a strange way, that they continue out there as a proclamation that "We were here" sometimes gives me a strange peace of mind about my own mortality.
> I feel this is one of those moments when the technical term "whew" is appropriate.
I thought mostly of "wow". They had a plan for a case when this "shout" didn't work. So they had at least two plans what could be done and the first one worked all right.
Half a century and 20 billion of kilometres from Earth and it still works.
I love the engineering behind Voyager 2. Is there a good book or documentary that folks here on HN recommend to go deep on the various engineering pieces behind Voyager ?
There's a book on the computers used at NASA from the beginning through to when it was written - late 80s, I think. Whole thing is available online [1]
It has a chapter titled "Voyager - The Flying Computer Center". [2] It gives a high-level overview of the computers and software. Three different processors, each dual redundant. 18 and 16 bit machines. Comparable to early 1970s minicomputers.
There's a good talk about the computers on the Voyagers available online here. [3]
As far as I know, beyond what's available between those two sources, very little otherwise is available publicly on the computer hardware itself - no detailed architecture descriptions, instruction sets, electronics details, etc. And no software listings. Though if I had to guess the 18-bit machines are a lot like - but not the same as - the OBP/AOP/NSSC series [4].
A bit of Voyager trivia: the computers were reprogrammed in-flight to give new abilities the Voyagers didn't have at launch, such as new image compression algorithms to allow more images to be returned than originally anticipated.
It's not particularly "deep", but The Farthest [1] is a nicely constructed documentary of the Voyager program featuring interviews with many of the relevant people.
I just a have a thought that in the very very distant future humanity will be able to do interstellar travels and they might detect and capture still travelling voyger 1 or 2 and learn what we was like back then tell to their kids the story about us. According to NASA - "It will take at least 40,000 years before either spacecraft approaches another star".
Perhaps you know this already, but Voyager 2 actually uses an tape recorder for data storage. I assume that all other options were either too fragile or too expensive back when it was built.
The vehicle records flight and sensor data to tape, and then periodically broadcasts that back to earth. As of 2022 the tape drive appears to be still working [1].
I believe tape was chosen mostly for capacity reasons. The Voyager DTR can store 109 megabytes, which was impractically large for any other storage medium.
Amazing it happened at all. Some scientists determined that 1977 allowed a once-in-a-millennia planetary alignment and pushed to get NASA money to send a prob. It was touch and go as to whether the funds would be allocated and if there would even be time to get it into the proper trajectory in time.
I wonder if current technology can achieve the same level of reliability.
For software, I fear complexity would make it much harder to build something that works for 50+ years, the probability of some nasty long term overflow bug or slow memory leak would make that difficult.
>I wonder if current technology can achieve the same level of reliability.
Yes, although that is very rarely a design goal.
>For software, I fear complexity would make it much harder to build something that works for 50+ years, the probability of some nasty long term overflow bug or slow memory leak would make that difficult.
Not that hard to handle. You just need to setup an appropriate architecture. In Aeronautic software systems you often have forced resets, if the software is not reaching certain points of exrcution.
You also do not have memory leaks, since you do not have dynamic memory.
Keeping an eye on CSIRO's Tidbinbilla website might yield some info as they get around to writing things up. It's the "horse's mouth" so to speak, as all Voyager communications goes via Tidbinbilla.
So basically exceeding the transmitter's normal operating range in hopes that Voyager 2 can hear it somehow. I'm guessing that between possibly destroying that transmitter (and having an expensive repair/rebuilding bill) and never hearing V2 again they've chose the first option. It honestly puts me in doubt on the chances that the self-realignment will work despite knowing that NASA have tested than mechanism more than what would a normal manufacturer have done.
I don't think anything is being exceeded. I have no idea of the actual numbers. but the calculation is something like. using the 100 meter dish we need to transmit at 30 kilowatts to make sure the signal can be received by voyeger2.
voyeger2 high gain is 2 degrees off, (do some black magic RF math) we need to transmit at 60 kilowatts to make sure the signal is received. good thing this transmitter is capable of 5 megawatts of radiative power.
Now I just made up those numbers. but the cool thing about nasa is how open they are. both when things go wrong "whoops we misaligned voyeger 2" and with specs, it probably won't take me long at all to find actual numbers.
> The probe would have reoriented itself towards the Earth in October, it does so twice a year.
I know that's what they say, but does anyone have any more details about that, too?
It's hard for me to reconcile the idea that the probe can automatically re-orient itself with the idea that it somehow got accidentally pointed in the wrong direction. You'd think they'd end every set of commands with instructions to check orientation and fix it if it's wrong.
Do any probes publish protocols that have distinct receive or telemetry protocols? Would be cool to be able to listen in to some of the closer ones, if possible.
Not sure about a mission as old as voyager, but most modern missions use CCSDS protocols, which are open and available online.
But just because you know the protocol dues not mean you have enough information to send a valid command. The commands are wrapped in a common protocol, but the commands themselves are typically mission-specific and definitely not made public.
Here’s an example of someone unaffiliated with the mission decoding JWST telemetry. While they were able to identify the packets defined per CCSDS protocols, they do not know the actual content of the packets, which are mission specific.
Thanks for the information. It's more just having fun decoding some telemetry protocol and maybe making some interesting graphs. Like maybe there's some interesting things they don't mind you knowing.
> And as someone else mentioned, the biggest hurdle with getting a command to voyager is access to a 70 m dish. Not many of those floating around…
So can we build things like this anymore? I am lucky if my cell phone lasts more than a year. I have a feeling that even if today's hardware manages to last 50 years, all knowledge of how to talk to it will be long lost. I am sure some of the code/firmware will be proprietary and then contractor will go out of business and that is that.
You're comparing the absolute pinnacle of human ingenuity and engineering from the 70's to a consumer-grade item in the 2020's.
We definitely have the capacity to build spacecraft to the quality standards of Voyager. Samsung would probably be willing to support your 2023 era Galaxy cell phone, if you're willing to spend $800m on it.
So can we build things like this anymore?
Of course we can.
I am lucky if my cell phone lasts more than a year.
That's definitely you, not the phone. Most people's phones last longer. Partially evidenced by folks complaining about updates stopping after a few years.
I have a feeling that even if today's hardware manages to last 50 years, all knowledge of how to talk to it will be long lost. I am sure some of the code/firmware will be proprietary and then contractor will go out of business and that is that.
Most of today's hardware isn't expected to last that long - and it wouldn't matter if much of it did. No one wants 50-year-old medical equipment unless it hasn't improved in 50 years. A 25-year-old computer isn't worth much to the average consumer. And some stuff will just be outdated: Do you really want to use one of the digital typewriters that combine a small-screen simple word processor with a typewriter?
Things that we build for longer-term science generally is built to a different standard than the stuff we use at home and has different constraints and different sorts of moving parts. Especially when considering extreme environments. And most of these are made by governments, universities, and other institutions that tend to have a long lifespan when compared to corporations. And when they aren't, we can make sure to make laws so others can use science research things even if the company closes.
Additionally: I think everyone expected Voyager to die long ago. It is akin to a fridge that happens to still be running 50 years later. A lucky accident.
> Additionally: I think everyone expected Voyager to die long ago. It is akin to a fridge that happens to still be running 50 years later. A lucky accident.
This speaks more to the quality of engineering than lucky accidents.
NASA sets high bars because many of these projects have a single chance of working. They make decisions that maximize the chances that the equipment will function within mission parameters, and the result is often something that far exceeds the original mission. See: Mars helicopter.
In the consumer space, an old fridge continuing to operate isn’t chance either. This truly is a case of “they don’t make them like they used to”.
Edit: I’m a bit surprised this is controversial. Curious why people feel this is wrong.
I have this condition called a recurrent corneal abrasion. Basically, my eyelid occasionally adheres to the skin on my eyeball and rips it open when I open them too quickly after waking up. It is very painful. The only thing that helps is an ointment that I put in my eye before I go to sleep. It comes in a tiny little tube, costs $40, lasts about a month. The last few tubes, I've kept even though they were "empty"; thank the hoarder in me. This turned out to be a saving grace, as the ointment is often sold out at nearby pharmacies, and I can't afford a new tube a month, anyhow. There tends to be enough left in each, if I spend a while squeezing the shit out of it, to get a small drop (1/10th of the recommended dose) into each eye for a couple of weeks. That's a few more weeks of not waking up feeling like someone has taken a razor blade to my cornea. I'll take it.
Also something about Cuba. I think my point is that we're a bit too cavalier with used electronics these days. The embodied energy and materials are huge, and we just... throw them out, in many cases. I'd like us to get used to repurposing these resources before scarcity forces us to. Really squeezing every bit of use out of them that we can. Bonus: because it's not the Hot New Thing Someone Is Trying To Sell You, you'll likely be less subject to the Hot New Trend Of Actually Being The Product.
And, for what it's worth: an original Apple I just sold for almost $450k at auction. Someone wanted that.
I guess, if the authentication / DRM server hasn't shut down long ago then, the certificates will have long expired. (BTW, what about a deep space DRM server for that purpose, somewhere around Jupiter? You can't run an asteroid miner without proper subscriptions to enable specific features!) ;-)
I have to say, it gets really old seeing this sentiment on HN anytime a post about older technology gets made. We absolutely can and do make technology designed to last this long - it’s just that the vast majority of us don’t work on anything with those kinds of requirements or priorities. There’s no reason you need to be building your CRUD API, data pipeline, or cell phone to last 50+ years and doing so would probably be actively harmful to the actual goals of those projects
For what it's worth, I actually agree. See: Curiosity or Ingenuity, both relatively new and far outperforming their expected life. Not to mention that I'm sure the ground-based transmitters that were used to actually fix Voyager 2 are likely not old, so it really isn't a case of "old is good" which helped in this case.
My comment is unfortunately an example where I let my late night shit-posting bleed over into HN, which is always regretted in the morning. Apologies ;)
Modern space electronics cannot last this long. It would require a significant effort to design similar hardware and have it fabbed today. Modern missions want capabilities that necessitate high transistor counts and that forces feature sizes that will never be robust against radiation effects over long durations.
Cassini lasted 20 years (1997-2017) and was functioning fine all along. It was intentionally ended, but there had been talk of sending it on another 20 year cruise to Uranus.
New Horizons is similar, now at 18 years since launch and still counting.
They're not quite 50 yet, but there's no reason to think either couldn't make it that long.
I'd realy like to se a post mortem about how that happened. I'd expect that all commands would be validated in a simulator with checks that mission parameters are not violated.
Imagining Blind Willie Johnson's Dark was the Night playing in background while Voyager drifting through the infinite interstellar void, reaching further humanity has ever reached, still gives me goosebumps to this day.
Does anybody know the bit rates, bandwidth, spreading, coding, and packet sizes of the messages they send? It must take several hours to send a message. I also wonder what the minimum number of bits they need to send to recover control is.
It is slightly amusing how the text says multiple times that the spacecraft "remains on its expected trajectory". Were they anticipating autonomous course changes?
This answers a question I had when they first announced the problem: how did they know it was 2 degrees off? Answer: because it was their mistake that threw it off, so they knew precisely where it was pointing.
I was born in the early 70's and the Voyager's have been travelling most of my life. In some ways they're like a touch-stone, something I always click on when I see an article.
One day, like me, they'll finally die, but until then may they travel onwards and outwards.
A tip of the hat to folks at NASA and JPL that built this marvel, with 70's tech, that still continues to amaze and delight us.