Man, that thing keeps chugging! Great news for JAXA, I wonder what additional science they're getting out of it. I know that we would have used any extra time for further downlink. So much data, such little bandwidth!
Probably environmental data, mainly temperature, and of course, they get to see how it progressively fails. They're starting to see some temperature sensor and battery issues for instance.
I took a look at the press releases and I can't see a lot of information about payloads, but if they're operations were anything like Nova-C (which I worked on), then they generated lots more data on landing than they could downlink at the time.
I would suggest main value is in seeing how this thing still lives! It was designed for one lunar day and has survived 3. So for learning about what still works and what doesn't this is invaluable for future reliability engineering.
In this case there does not appear to be any public evidence that they did much testing in below freezing temperatures. This isn't to say that they couldn't do the testing, just that they didn't do it because the mission parameters did not require it. There don't seem to have been many "ongoing experiment" type sensors on the lander which would be significantly useful to keep going longer than the initial mission.
As for radiation, I attended a lecture series last year about the various kinds of radiation testing involved in designing hardware for satellites, they probably did do similar things to those for satellite electronics (stuff like firing beams of IIRC electrons over the PCBs to see what parts might need reinforcement/redundancy against random charged particle strikes).
Part of the unknown is likely what temperatures, how much radiation, of what kinds. The SLIM lander can land in places previous automatic landing systems couldn't, so it's very possible this area has never had such measurements done. There's still much we don't know about the moon, including how cold the lunar night actually is in different areas.
Also, the point of SLIM was to test the new automatic landing system, a combination of camera-based navigation and a new landing strut system meant to handle inclines. 100 meter precision instead of previous 20 km. Everything that comes after that is near-free extra data.
I recently watched a documentary about the ongoing Voyager missions[1], and this same tactic (turning instrument heaters off to save power, hoping the instruments still produce useful data) continues to be used with success.
In that documentary, it was remarked by one of the engineers that scientists often designed these instruments to be more robust than what was expected.
JAXA said that the mission design for the lander was one lunar day, which generally doesn't involve dropping much below 0C (https://space.stackexchange.com/a/37374)
So ya, they probably didn't do any significant testing below 0C.
They seem to reference a paper at "19)"[0] which goes into much further detail. It's odd though reading where they reference it, they talk about nothing below 0 C at all.
Is that because that's the limit of the sensor or an actual reading. It's something that happened with dosimeters at Chernobyl where people didn't realize the levels were so bad until someone pointed out that the sensors were maxed out because the levels were higher than the range the units were designed to display. I would hope that's not the reason, but it has happened in the past
Most materials contract when they cool down. For example, differential contraction between metal contacts and semiconductor materials can cause them to detach or break, disrupting electrical connections. Also extremely low temperatures can lead to changes in semiconductor properties such as carrier mobility, which affects how efficiently electronic signals are processed.
Couldn't all connectors be made with some kind of expansion joint, like how infrastructure (bridges etc) are made? Given how they use older nodes, this may perhaps even be possible for transistors (Assuming they also have such expansions)?
Normally materials in chips are selected so that they have similar thermal expansion coefficients - otherwise they would fall apart just from reaching normal work temperatures.
There's another problem here: below a certain temperature semiconductors become insulators. You're running the risk of your chip shutting down in a disorderly manner.
Of course you can do all sorts of things to make the electronics able to survive, but ultimately that supposedly just wasn't part of the initial design requirements they settled on.
It's good to remember that the total lifetime budget of this lander is only $121.5 million, and that includes all the staff still receiving data. It was always intended to be a technology demonstrator, not an ongoing science lab.
I would guess the main value is in practicing the technological skills of sending a thing to the moon and maintaining connection with it.
Are we able to also simulate the radiation in a laboratory?
I will wager that this was done prior to launch, and current longevity is not a complete surprise.
As for radiation, I attended a lecture series last year about the various kinds of radiation testing involved in designing hardware for satellites, they probably did do similar things to those for satellite electronics (stuff like firing beams of IIRC electrons over the PCBs to see what parts might need reinforcement/redundancy against random charged particle strikes).
Also, the point of SLIM was to test the new automatic landing system, a combination of camera-based navigation and a new landing strut system meant to handle inclines. 100 meter precision instead of previous 20 km. Everything that comes after that is near-free extra data.
In that documentary, it was remarked by one of the engineers that scientists often designed these instruments to be more robust than what was expected.
[1]: https://www.itsquieterfilm.com/
https://www.jstage.jst.go.jp/article/tjsass/66/6/66_T-22-48/...
So ya, they probably didn't do any significant testing below 0C.
[0] https://ieeexplore.ieee.org/document/1145732
There's another problem here: below a certain temperature semiconductors become insulators. You're running the risk of your chip shutting down in a disorderly manner.
−273,15°C < −247°C.