33 replies
1d12h
I continue to be fascinated by how they:
1. Are able to diagnose the problems remotely at such distances and on such old hardware. How can they even measure the thruster tube apertures here?
2. Decide what actions to take. It's not like they have a local test device to experiment on is it? (Even if they did I can't imagine how they'd reproduce the conditions of the real thing.) And if they choose poorly, I'd assume the mission's over. There's no replacing Voyager 1 if they brick it.
3. Have such fine control over the hardware. For something built in the 70's when RAM was largely measured in kB, they seem to have an insane amount of flexibility to remotely reconfigure the equipment. Whatever they did, there must have been some real foresight.
In short, the Voyager spacecraft were designed (after considerable design and operational experience from Mariner and Pioneer spacecraft) to be able to operate for a long time largely automatically without too much hand-holding. All major systems are multiply redundant and may be remotely turned on and off. While there has been personnel turnover on the program, it has not been of a magnitude to jeopardize program continuation. Finally, program management has been media savvy and well politically connected ensuring that operations are still funded. (Contrast with other missions such as Magellan to Venus which was deorbited while it still had propellant reserve leaving some portions of the planet unmapped)
JPL has (or had?) a "retiree badge" program that permitted retired staff to continue to access their office. Many programs benefited from highly knowledgeable personnel essentially continuing to report to the office every day without pay (not being paid comes with the luxury of not having to worry about being laid off if your charge accounts don’t have enough funds!) It was an absolute privilege to learn from these people.
So, culture preservation is important for success of highly technical endeavours. Don't tell it to your run of the mill MBA.
but if someone is paid $0 it must mean they are worthless, surely the excel file doesnt lie
No, you need to think like an MBA.
It means that they are the most valuable employees, because their productivity per dollar is infinite.
It also means that when you sort all the (non-executive, of course!) employees by total comp in preparation for layoffs to make the budget look better, they are at the bottom of the list.
oh yes, you are absolutely right :D
Run off the mill MBAs are not allowed anywhere near to stuff as important as Voyager.
All the web apps, that run of the mill devs are implementing are nowhere near as important as voyager.
I think we are fine ;)
It also helps that JPL is a public good and not a wealth-extraction machine
I think it is inevitable that organizational "culture" changes. The tricky part is figuring out exactly what parts need change and what parts don’t.
For example JPL used to have beauty pageants ("Miss Guided Missile"). More recently, management appears to be trying to adopt policies and procedures from some venture funded commercial space companies. It is not clear how helpful these efforts will be given that these organizations are in fundamentally different businesses.
https://www.jpl.nasa.gov/images/slice-of-history-70th-annive...
I guarantee you that a run of the mill MBA program will say something about institutional knowledge being important. You’re needlessly and unjustifiably finding an excuse to dunk on something that you know people here dislike, despite it having very little to do with the actual topic. Which…fine. But you’re also so keen to do it that you aren’t even correct. Please stop projecting every gripe you have with your boss onto this mythical “evil MBA”.
Any idea how many people took advantage of the retiree badge program, or any individual people who continued to put in substantial hours?
We do something similar at my job for the greybeards that were very influential on our current projects. They "retire" but are retained as a very part time employee that only get paid if we need to bring them in for a day or two to help answer some questions. The managers love it because they don't need to find work for this person, pay for any benefits, or need to get approvals to pay them like contractors. As long as we keep doing relevant work to their expertise, they will continue being retained. There is a limit for how many hours they can put in because these people are incredibly expensive as they usually retire at the top engineer level and retain the equivalent hourly wage for that position.
Paying someone as sporadically as you would a contractor but they’re only getting their usual hourly wage? Doesn’t sound expensive to me. Sounds like a ridiculously good deal for the employer.
Yeah! Being almost 50 years old it's not like people is not working there anymore but that probably a bunch of people in the original project has already died!
Great forward thinking
What blows my mind is the organizational knowledge needed after so many years to keep it going. You don't just hand a guy some man pages when he comes onto the project. I'm sure people have aged out, yet they still understand the complexities in the design. That is something we need to understand and prioritize in the systems we build today.
Startup: "Do you remember when we inserted this quirky module running in AWS? We can use that to implement this next feature. That was a useful decision!"
Voyager: "Do you remember when our parents inserted this quirky module that has since left the solar system? We can use that to turn it off and on again. That was a mission critical decision!"
that aws module isn't supported anymore though, we need to npm install half the internet for the 2 line library that replaces it.
Not parents for most of us. Grand or great-grandparents. The senior engineers were highly educated and experienced, implying ages 30-50s during development. They are in their 80-100s now.
When Voyager failed last year with a CMOS memory error, one of the big problems was that a bunch of low level information was gone or conflicting. For example, they sometimes had to guess assembler instructions because the code printouts were low quality photocopied pages. Or because there were handwritten comments or comments scratched out with pencil without any clue about why it was done.
One saving grace was the fact that the architecture and the code space was simple enough so that they could reason through the symptoms and actions to take, something that would have been much harder with modern spacecraft.
Check out this amazing talk: https://youtu.be/dF_9YcehCZo?si=W_b3NJ7vgxaYS1__
This thing hits me. True.
I would imagine if the design/assembly information was broadly available (internally) in the past, there's probably one or several "digital twin" emulations of the craft, or at the very least specific subsystems of it's computing resources. There must be some kind of analog/simulation of it's software just for proving "bugfixes" before upload, like the coms error and subsequent setting of the "solar system record" for "furthest distance remote code update" earlier this year.
There isn’t a simulator or digital twin for voyager. It has a bespoke processor made with 74* style logic. One guy will puts together a command and they will have a review where the other engineers will try and independently verify it. Then they copy and paste the command somewhere to “run it”. It happened, fairly recently, that the command had a typo that was caught in review, but the “wrong” pre-review command was used and the attitude became off by so much that they lost contact. It was only by cranking up the power at Goldstone that they got a command through. This fundamentally changed their understanding of the largest angle for which they could still communicate with the spacecraft. They just hadn’t wanted to try larger angles before because it was too risky.
That's the great thing about such a simple design, you can actually sit down with pen and paper and verify operations.
Forget about a digital twin, they don’t even have an assembler for the CPUs. It’s all hex values.
And quite recently they also recovered from a faulty ram module, with a radio link speed that is maybe below the kbit/s now !
https://science.nasa.gov/mission/voyager/spacecraft/
The data playbacks were initially transmitted at 7200 bps, but this was dropped to 1400 bps in 2007 (https://voyager.gsfc.nasa.gov/Library/DeepCommo_Chapter3--14... - page 72).
RTT ~46 hours (one-way light time 22:49:59, per https://science.nasa.gov/mission/voyager/where-are-they-now/)
And you should see the ping times.
I suspect it’s far lower than that.
There's a documentary "It's Quieter in the Twilight" about the team keeping the Voyager missions alive: https://www.youtube.com/watch?v=8vJT8AW0wYw
Included with Amazon Prime in Canada
Looks very interesting , thanks for sharing
I would bet they have one or more simulators (“digital twin” in the parlance of our times). I’d want one simulator to always reflect the current state of the probe (with state data assimilated periodically from the real probe). Then other simulators can be used to test management changes to see how the system might respond.
Sadly they do not, but they are starting to write one.
You should totally check out the talk from a few weeks ago that I link to in my other comment. It answers your 3 questions and more.