FWIW I found the report's summary (a few days ago when this was also discussed) essential for clarifying for me what the linked "article" was talking about regarding "anode overhang" and "edge alignment".
To summarize it myself, the battery is cylindrically wrapped sheets of anode/cathode material, and you want the top edge to have a straight line consistently having the anode layers (|) sticking out beyond the cathode layers ()
edit: Welp, the unicode I illustrated with got stripped out; so trying again with pipes
anode
cathode
a a
c c
| | |
||||||
The pics with red and blue dots didn't immediately make this obvious to me, but made sense once I understood the "overhang" meaning.
I find their table of advertised vs actual capacity to be misleadingly negative. They only discharged to 3v. 2.7 could be viewed as more standard. 2.5 is not unheard of. For example, the vapvell 4000 they said was around 3000. They even have a note that says that isn't a reasonable capacity estimate. And yet they still put the percentage and the number. As if they've falsely advertised. However, if you go to vapcell's graphs on their website, it all tracks. Feels out of scope for the report and shouldn't have been done
How do you even drain a 3.7V lithium ion battery below 3.3V? My devices that use 18650s will not let them go below that. Is it 3.3V nominal and the actual voltage is lower, like how they’re 4.2V fully charged?
After we monitored many battery fires, we decided to build a casing which can sustain lithium fires for typical e-bike batteries: https://www.youtube.com/watch?v=v0NXXfCA2CY
I just got a new tenant in my rental house and this time I had to add a clause to the lease enumerating brands of lithium-ion batteries the tenant is allowed to keep on the premises. These things are basically grenades that you can buy for $2 on Temu. Stronger regulations needed.
Your best predictor of cell failure is brand + where you sourced it. Good brand cells sourced through a reliable channel are pretty much impossible to get a failure out of without gross abuse of the cells.
But far more dangerous than single cell failure is pack failure and there the quality of the welding and the balance wires is probably a much bigger factor than bad quality cells. You can easily create a nice looking bomb from grade A cells, and even manufacturers with good name recognition will mess this up from time to time.
So buy packs made with quality cells and ensure that whoever makes the pack has proper welding gear and QA in place to avoid surprises. Never ever use cells without a BMS unless they are in a 1S configuration.
Cells are subject to aggressive binning – akin to the early days of MOS process. And specs and process capability adhesion in the industry are not as "serious" as they are in the modern, U.S.-led semiconductor industry.
Source: I work on 100% Si anode batteries constructed in part with a litho-derived laser process at Enovix.
Summary Statement: He summarizes the results of the "sabotaging" attempts by saying, "we just couldn't for the life of ourselves get a nonprotected modern lithium battery to do anything" [02:40:27], in terms of causing a fire. (referring to 18650s)
There are certainly problems out there with cells but it was a surprising statement from someone with a ton of real-world experience. Also they are a very conscientious company so they don't deal with dodgy stuff if they can avoid it.
problem with Lumafield is their pricing. Last time I checked with them. They don't sell the CT machine. It was a lease/subscription at yearly cost of $75,000. It was not justified for what we are doing.
I get they won't disappear entirely, but what are the applications that 200-225 wh/kg LFP and forthcoming 175 wh/kg sodium ion can't handle with their much better safety profiles?
For higher density applications, semi-solid state is coming for nickel/cobalt chemistry lithium ion, and it has much better fire safety.
Just a few days ago the apartment above ours burned down alledegly because of an explosion of an ebike battery. They left it charging overnight and exploded in the morning. Fortunately nobody got hurt, but it was really scary experience.
I'm not sure who is in charge of advertising at lumafield but in the last few weeks I've seen James Hoffman use lumafield CR scans on coffee pucks, Jeff Geerling use them on a Pi 500+, and now this. I'm probably not their target market but I definitely know who they are now.
sorry for a dumb question but are old dead cells still dangerous just sitting around?
I discovered an ancient powerbank the size of a deck of cards that I had not charged in many years, was blown up like a football (imagine an a sealed envelope but full of air)
So that's a dead cell which shouldn't have a charge left but I guess the chemistry is releasing gas?
Scared me, wrapped it in aluminum foil and put it outside asap
If the cell is fully discharged, there isn't much danger because all of the energy that _would_ cause an explosion/fire is gone.
However, a LiPo cell that has entered Spicy Pillow mode (empty or not) can still theoretically burst, leaking or spraying toxic materials everywhere. So you were right to put it outside. Make sure you recycle it appropriately, so that it doesn't go into a landfill and poison the drinking water of our grandchildren someday.
Note that batteries that catch on fire are a form of what AI safety researchers would call unaligned AI. Incentive structures that lead to the production of unaligned dumb AI appear to be quite an under researched blind spot in AI safety research.
> All 33 of the batteries with the [negative anode overhang] defects came from the 424 sold by low-cost brands or brands selling counterfeits.
>
> …
>
> None of the name brand OEM batteries were found to have any problems, and while there was some deviation in anode overhang for rewrap batteries, the deviations were seven times worse for the low-end cells.
Based on those stats, it sure sounds like the following is happening:
1. The reputable OEMs (Samsung, Panasonic, Murata in this study) do adequate quality control for defective batteries. The low-cost OEMs do not do adequate quality control.
2. Some portion of the defective or not-fully-tested batteries from the reputable OEMs get "diverted" to rewrappers, who essentially just relabel the batteries without doing any further testing.
22 comments
[ 6.3 ms ] story [ 46.9 ms ] threadhttps://7802750.fs1.hubspotusercontent-na1.net/hubfs/7802750...
To summarize it myself, the battery is cylindrically wrapped sheets of anode/cathode material, and you want the top edge to have a straight line consistently having the anode layers (|) sticking out beyond the cathode layers ()
edit: Welp, the unicode I illustrated with got stripped out; so trying again with pipes
The pics with red and blue dots didn't immediately make this obvious to me, but made sense once I understood the "overhang" meaning.But far more dangerous than single cell failure is pack failure and there the quality of the welding and the balance wires is probably a much bigger factor than bad quality cells. You can easily create a nice looking bomb from grade A cells, and even manufacturers with good name recognition will mess this up from time to time.
So buy packs made with quality cells and ensure that whoever makes the pack has proper welding gear and QA in place to avoid surprises. Never ever use cells without a BMS unless they are in a 1S configuration.
Sounds like a bunch of amazon merchants. Imagine buying lithium ion batteries from a vendor with "trust" and "fire" in their name at the same time.
Source: I work on 100% Si anode batteries constructed in part with a litho-derived laser process at Enovix.
https://www.youtube.com/watch?v=j92Gt4VviSQ
Summary Statement: He summarizes the results of the "sabotaging" attempts by saying, "we just couldn't for the life of ourselves get a nonprotected modern lithium battery to do anything" [02:40:27], in terms of causing a fire. (referring to 18650s)
There are certainly problems out there with cells but it was a surprising statement from someone with a ton of real-world experience. Also they are a very conscientious company so they don't deal with dodgy stuff if they can avoid it.
For higher density applications, semi-solid state is coming for nickel/cobalt chemistry lithium ion, and it has much better fire safety.
I discovered an ancient powerbank the size of a deck of cards that I had not charged in many years, was blown up like a football (imagine an a sealed envelope but full of air)
So that's a dead cell which shouldn't have a charge left but I guess the chemistry is releasing gas?
Scared me, wrapped it in aluminum foil and put it outside asap
However, a LiPo cell that has entered Spicy Pillow mode (empty or not) can still theoretically burst, leaking or spraying toxic materials everywhere. So you were right to put it outside. Make sure you recycle it appropriately, so that it doesn't go into a landfill and poison the drinking water of our grandchildren someday.
Based on those stats, it sure sounds like the following is happening:
1. The reputable OEMs (Samsung, Panasonic, Murata in this study) do adequate quality control for defective batteries. The low-cost OEMs do not do adequate quality control. 2. Some portion of the defective or not-fully-tested batteries from the reputable OEMs get "diverted" to rewrappers, who essentially just relabel the batteries without doing any further testing.
Is that a correct read?