Li-ion roars can predict early battery failure, MIT boffins say When lithium-ion batteries degrade, they emit acoustic signals that reveal what's going wrong inside. Now, MIT researchers say they've figured out how to interpret those sounds, and the subtle creaks and pops that come before major failures, to help predict problems before things go up in smoke. Led by MIT chemical engineering …
Battery: "What is my purpose?"
Scientist: "You provide energy to an incredibly powerful handheld device capable of uniting people across vast distances and sharing infinite knowledge but that actually only gets used for porn, memes of cats, and trolling other people... sometimes all at the same time."
Battery: "Oh my god!"
Scientist: "Fine, you can spend your existence passing the butter instead. *belch*"
One more horror of Cyber-conversion, when you feel your battery starting to fail: "Pain!" | World Enough And Time | Doctor Who
I'd love to see this in on-grid home batteries. I love the idea of the tech, and it would solve a lot of issues with current electrical infrastructure, but i've been super wary of having a huge Li-ion battery inside my home. If this tech gets reliable enough to catch most failures before they happen, it would come a long way to convince me to actually use this tech.
And why are you so scared of a well managed battery?
They are usually LFP chemistry, rather than NMC - and that change alone make a significant difference. They are also coupled with a decent BMS, because when you have a few grand of battery it's worth a little extra electronics to make sure it works effectively for as long as possible (and fire tends to abruptly shorten the lifespan of battery)
There are very few reports of home storage battery fires, and I mean *very* few:
https://assets.publishing.service.gov.uk/media/5f761b828fa8f55e33275cfc/domestic-battery-energy-storage-systems.pdf (section 5)
I'm far more concerned about my phone, or a small transport battery (e-scooter, e-bike, wheelchair power add on) than I would about a home storage battery. Even vapes seem to be a more significant risk... And I don't particularly worry about those (but then I don't go for the cheapest disposable battery available, and no-one in the house vapes).
"How do they set up a double-blind test?"
That methodology wouldn't work. Loads of sensors would need to be placed on loads of battery packs to where there was a big enough statistical universe to make some qualified predictions. Sorta like they do with volcanoes. Nothing, nothing, nothing, RUN!! FIRE!!
I was watching a cat video..........
The cats in the house were watching the wall around the fireplace with rapt attention. The hooman was sitting on a couch opposite the fireplace. The cats all bolted and the stone glued to the wall all came off in one go.
If cats could be trained, maybe they'd be able to hear a battery getting ready to go froush and give a warning.
Last century I was tasked by a Jet Heritage charity to identify when an engine was starting to fail, rather than just using the manufacturers stated number of hours.
First of all we got it written into the contract that this was on a best try basis and unexpected catastrophic events could still happen without warning!
What follows is a simplification.
I built up a profile of all the jet engines they had during normal starting cycles using the energy profile required by the starter and the typical sound it would make.
This was then built upon with each subsequent start of each engine.
Any sudden change in either profile, or a gradual drifting out of a standard deviation over all the engines, strongly suggested that the bearings (the only moving part in a jet engine) were getting worn and needed looking at.
It saved them a whole lot of money compared with just swapping out on number of hours use and to my knowledge did not result in any Rapid Unscheduled Disassembly's. (And yes, that term was in use then!)
