5 ms·
CT scans of 1k lithium-ion batteries show quality risks in inexpensive cells
- progbits 1y agoTo save everyone the hassle of filling out the "get the report" form: https://7802750.fs1.hubspotusercontent-na1.net/hubfs/7802750/White%20Papers%20and%20Guides/Batteries/Lumafield%20Battery%20Report.pdf https://7802750.fs1.hubspotusercontent-na1.net/hubfs/7802750...
- nom 1y agothank you
- gertlex 1y agoFWIW 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.
- MBCook 1y agoSo the big danger here is an internal short being far more likely?
- gertlex 1y agoThat was my general sense. And that "nominal" is really easy to identify visually with a (relatively slow) scan. I'd guess that means it's a clear quality control thing you (i.e. battery pack makers who are buying the cells) can expect the better manufacturers to do, and pay a slight premium for.
- dpedu 1y agoI seem to recall this being a factor in the Samsung Galaxy Note 7 battery fire debacle. The article linked below has an xray image where you can see the anode/cathode overhanging and then being shorted by subsequent damage. https://www.theguardian.com/technology/2017/jan/23/samsung-blames-faulty-batteries-for-causing-galaxy-note-7-fires#img-2 https://www.theguardian.com/technology/2017/jan/23/samsung-b...
- stavros 1y agoIs this what's shown in the photo here? https://imgz.org/i55rjGiq/ https://imgz.org/i55rjGiq/
- gertlex 1y agoYep! where the green arrow is pointing is what I was ascii-lustrating
- Neywiny 1y agoI 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
- ricardobeat 1y agoIt even says "Conservative discharge that would not extract maximum performance". A lot of electronics built for li-ion will not work <3V so it's a perfectly reasonable benchmark, especially to show the difference between brands.
- londons_explore 1y agoA lot of electronics need a 3.3v rail and produce that from a linear or buck regulator, so won't work below 3.4 volts.
- Tade0 1y agoBattery operated devices, while nominally 3.3V, commonly work down to 3V or even 2.7V in line with the voltage curve of lithium (not li-ion) coin batteries.
- edgineer 1y agoI went to vapcell's website, looked at their charts and specifications. I disagree with your assessment. They specify 3800mAh minimum discharge at 0.2C for their 4000mAh cell. They don't show a discharge curve at 0.2C in their charts (would be 0.8A) but they do have a 1A discharge. When the 1A discharge curve reaches 3V, the energy discharged is right around 3800mAh. Lumafield discharged at 0.2C, and they saw only 3055mAh. Vapcell's site mostly shows high current discharge curves, where yes there is more capacity below 3V available, but with Li-ion at lower currents, the curve is very steep past 3V, not much more capacity left after that. And when you're discharging at high current you won't expect to get the full capacity, anyway. I'll also take this time to point out lygte-info which is a treasure trove of battery tests. [0] https://www.vapcelltech.com/h-pd-193.html https://www.vapcelltech.com/h-pd-193.html [1] https://www.lygte-info.dk/ https://www.lygte-info.dk/
- oulipo2 1y agoAfter 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 https://www.youtube.com/watch?v=v0NXXfCA2CY
- djmips 1y agoThat's a great idea. Are you making a product or sharing the build?
- toomuchtodo 1y agohttps://infinite-battery.com/ https://infinite-battery.com/ (no affiliation)
- oulipo2 1y agoWe're building a product, but that's easy to adapt to many setup (we have many end-users mounting our battery on their e-bikes, for now we are compatible with most systems, Bafang, Brose, Bosch, etc, working on Shimano now)
- kragen 1y agoI think you mean "contain". Very nice.
- oulipo2 1y agoIndeed! Thanks :)
- jeffbee 1y agoNeat. How do bikes that have the battery entirely inside the metal downtube compare? Edited: Now that I looked at your product I have low expectations for an unbiased answer :-|
- oulipo2 1y agoWe're working on an internal downtube battery too! It's basically the same tech, the batteries work the same, it's just the form-factor which changes
- jeffbee 1y agoI 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.
- ReptileMan 1y agoWon't help much. Way too much cheap lithium ion crap with incapsulated batteries is flooding the markets
- jeffbee 1y agoIt does help. It raise the opportunity to discuss it, which is helpful because not everyone is aware that this is a problem, and like many other lease clauses it is there to assign blame in case an exceptionally bad event comes to pass.
- dylan604 1y agoNah, tenants will sign just to move in, and forget about this clause. Just like any tenant that smokes weed disregards no drug clauses. Tenants are going to do what tenants do. Unless you're doing routine walk throughs looking at every single thing in the place that could use a battery, then you'll never know. It's a useless "regulation", and we have plenty of those in the world already.
- jeffbee 1y agoIt's wrong to have an unjustified low opinions of tenants as a class. They are normal, well-meaning, intelligent people like any other.
- quesera 1y agos/tenants/humans/g The large majority of tenants do not read their lease paperwork carefully enough for this clause to make a difference. A large majority of humans don't read EULAs. A lease clause is useless for liability, because your landlord's insurance won't be able to shake down the tenant anyway. Someday soon enough, your insurance might require a clause like that. It will still be ignored by tenants though.
- DenisM 1y agoThe conclusion I drew from this is that if real capacity is much less than advertised the battery is best avoided.
- privatelypublic 1y agoThats a policy to have when buying things in general. Even more so when gram for gram it holds 1% the energy of gasoline... but doesn't need the help of oxygen, or any decernable trigger to make a hash of things.
- leptons 1y agoA common practice by shady sellers with 18650 lithium cells is to advertise a group of 4 of them as "2000mAh", but it's the total of each cell, not the individual cell ratings. So you're really getting four 500mAh batteries, which really sucks.
- jacquesm 1y agoYour 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.
- josephg 1y ago> 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. How do you put that advice into action? When I'm in an electronics store, they don't typically let you look at the welds inside each battery bank. (And if they did, I have no idea what I'd be looking out for.)
- jacquesm 1y agoGood point. It requires disassembly, and if you're not comfortable with that it is a lot harder to get a good idea on this. But there are plenty of videos and photos online of various disassembled packs from high volume manufacturers. Quality cells are a relatively small group of brands so that should be easy enough to check for, and you'll know looking at just one side of a cell group whether or not it was put together half decently or whether it is messy. Stuff to look for: loose balance wires, bad welds (welds that don't penetrate so they're loose), pointy edges of connecting strips that touch the plastic insulation of the cells, insufficient spacing between the cells, insufficient insulation near the welds. For one-off purchases this won't help, but if you are a reseller or larger purchaser then it definitely is a good idea to tear down a pack and inspect.
- askl 1y agoYour could take your CT scanner to the store to look at the cells.
- bob1029 1y ago> Efest, Vapcell, Trustfire, Treasurecase, Benkia, SOOCOOL, Maxiaeon 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.
- clumsysmurf 1y agoVapcell is very popular in the flashlight community for providing unusual sizes like 16340, 18350, 18500, 26800, etc. They re-wrap other manufacturer's cells (but don't disclose who). I dislike this practice because, along with not knowing who makes the cell, you lose the original mfg batch codes, etc. Sometimes they will simply rewrap something like a Samsung 50S. Those will be fine. But others ... who knows.
- malchow 1y agoCells 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.
- Liftyee 1y agoLast time I checked, most manufacturing of semiconductors isn't done in the US... sure, they are leading in design, but the manufacturing is elsewhere.
- xattt 1y agoI went down their rabbit hole, and a conventional tech CT is 10 hours??
- kg 1y agoMy understanding is that material composition can make a CT scan take a really long time. It makes sense to me that scanning a battery would be pretty slow, given what they're made out of.
- xattt 1y agoI just assumed it would be impractical due to physical changes of the object from multi-hour exposure to X-ray energy.
- kragen 1y agoI don't think ionizing atoms inside a battery will harm it. They don't have DNA.
- adwn 1y agoI don't know about batteries, but ionizing radiation can definitely permanently damage microelectronics, and those don't have DNA either.
- VectorLock 1y agoThankfully functionality isn't usually necessary to get a successful scan, unlike living targets.
- kragen 1y agoIt can, yes, but batteries also don't have microelectronics.
- habi 1y agoMetal objects don’t change that much due to the radiation.
- bahmboo 1y agoSeems like a good time to highlight this video from Grin Technologies and their experience with ebikes over the years. https://www.youtube.com/watch?v=j92Gt4VviSQ 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.
- bahmboo 1y agoIt's a long video but one takeaway is to only use cells from the top manufacturers: Samsung, Panasonic, LG and Sony. They are amazingly safe.
- eth0up 1y agoMolicell? I have one in 21700 and it seems a great battery so far. Otherwise I'm a Panasonic guy. I ask because it sits on my forehead, inside a headlamp. Having my hands tied, addressing a problem quickly might be difficult. And them rascals get toasty fast when they ignite.
- bahmboo 1y agoThe anecdata is Molicell is good stuff. Was going to include that but I have no references.
- lambda 1y agoYeah, Molicel are quality cells as well.
- fnordian_slip 1y agolygte-info.dk rates them quite highly iirc.
- jdietrich 1y ago
- ycui1986 1y agoproblem 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.
- viasfo 1y agoBut a conventional CT scanner costs $300k-$1m plus $50-100k per year for software licenses and maintenance contract. Lumafield’s basically giving you the entire system for just the annual software/support cost on an older-style machine. My company is considering one and it’s a more attractive model once you start comparing realistic costs.
- CamperBob2 1y agoI wonder if ultrasound techniques might provide an alternative. If it takes 10+ hours to do a CT inspection, as someone pointed out elsewhere (if I understood correctly), then that's a lot of DSP time. For that matter, jeez, how long does it take to just whip out a Dremel tool and take the battery apart for inspection? I must have misunderstood that comment.
- jonbruner 1y agoScan time depends on material composition in the object you're scanning and your requirements for resolution. You can scan a dense steel object overnight to capture micron-level detail, or you can scan a plastic object in a few seconds to search for a known issue like a crack. Battery scans are very fast; the scans in the report took less than a second. Total cycle time on a Triton CT scanner is under 5 seconds when you account for part handling.
- AtlasBarfed 1y agoI 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.
- kragen 1y agoDrones seem like the biggest one? Maybe some drones can cope with the extra weight.
- jdietrich 1y agoLFP is really catching up in terms of energy density, but not power density. LFP cells would give incredibly weak performance in something like a power tool battery, because they can't supply anywhere near the peak current of modern NMC cells, particularly pouch cells or tabless cylindrical cells. For bulk storage LFP is a no-brainer, but there are still a lot of applications where you really want fast charging or need to handle big surges of current.
- AtlasBarfed 1y agoSo semi-solid state then? What about sodium ion?
- gdotdesign 1y agoJust 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.
- withinboredom 1y agoThis is actually pretty common with e-bike batteries (as in fires per thousand when compared to other battery packs). Dunno why.
- dijit 1y agoExtremely large currents and a race to the bottom price (despite lithium-ion being expensive, people are eager to get “value” for the money which rewards conpanies which cut corners).
- PicassoCTs 1y ago[dead]
- fancyfredbot 1y agoI'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.
- cogman10 1y agoIt's pretty smart, IMO. They are targeting the nerds watching tech videos in their offtime. That's a great way to plant their name into the heads of said nerds when the boss decides they need a CT scanner for whatever reason.
- tredre3 1y agoAdvertising works. Shocking, isn't it?
- esseph 1y agoGood advertising works. Most advertising just makes it harder on everyone else, and things more expensive.
- ck2 1y agosorry 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
- bityard 1y agoIf 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.
- eightnoteight 1y agoseems obvious in hindsight, but never knew CT scans are used in manufacturing QC
- imtringued 1y agoNote 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.
- dxxvi 1y agoWhat about expensive cells? Any quality risks in them or they are perfect?
- leptons 1y ago"Buyer beware" always applies, no exceptions.
- dlenski 1y ago> 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. Is that a correct read?