Battleborn - hazard by design -

[EEVblog]

Will already put out a followup video where he looks at a couple other exmple batteries with fuses (and one which does not have a fuse, which he states would fail the test):

It's crazy they don't have a known rated fuse inside, why expose yourself to that kind of liability?
To save a few dollars on some crazy convoluted plastic DIY solution? 

One thing Will's fire video didn't have was measuring the short circuit current, that would have been handy to know.

[EEVblog]

They will be able to easily claim that he damaged the unit by cutting it open and bypassing the BMS

Yeah, "Lid was opened, so of course the heat that was supposed to trip the "fuse" could escape, test invalid" 

I'm sure they will use that excuse. It was also windy, providing a cooling function. And the external oxygen fueled the fire etc 

[tszaboo]

This is not an engineering problem. It is a legal problem so apply a legal framing, not an engineering framing.

Feel free to cite a legal reference showing that companies only have to get their products certified and not do anything outside of that.

"Unfortunately, the law is clear that strict liability does not apply to UL and establishing negligent inspections or procedures where these inspections take place halfway around the world is a formidable task."

https://corporate.findlaw.com/litigation-disputes/suing-underwriters-laboratories-in-a-products-liability-case.html
https://www.reddit.com/r/ebikes/comments/1j8zz6m/ul_certification_once_again/
https://incompliancemag.com/product-liability-law-and-its-effect-on-product-safety/

The only thing that an NRTL can "guarantee" is that particular example tested passed whatever the standard calls for regarding construction and test requirements. Not all standard have many construction requirements to go with the tests either, you have to read the whole thing and understand it.

Moreover, not all NRTLs are made the same: people talk about UL here but Intertek was the one used by the battery company, not UL. They're widely known to have creative interpretations of the standard and their testing is nothing like going to UL proper, generally it's the most clown-car type of NRTL you can choose for your product.

People that see the "UL" standard logo on a device think somehow that it's safe, while in reality it simply means that a lab tested one sample, using their interpretations of tests and construction requirements, and has then washed their hands of liability of any production units that come after it. The OEM is in the end wholly liable like they should be, even though a 3rd party listing provides some protective evidence for due diligence.

Not necessarily. We get visits from NRTLs 4 times a year where they inspect the safety related parts of our production.
I'm not 100% familiar with the US law, but under the EU law you are still liable to make sure the product that you ship is safe.
Also, just passing a standard, doesn't mean that a product is safe. Or passing a test.

In the case with the plastics in between the doubled nuts make some sense. The fist nut can not be torqued down fully and the part locking the nuts is the torque between the nuts. It is however rather tricky to torque correctly - want 3 hands for this.

I would be more concerned that the plastic would flow and creep over time. Making a loose connection, arcing or the contact resistance heating.

[jheissjr]

Could it be Battle Born said one day in an internal meeting regular car batteries don’t have internal fuses so our batteries don’t need them. Could they be thinking fuses are done on the customer side and in hindsight they now realize that’s incorrect.

[Freesurfer]

Could it be Battle Born said one day in an internal meeting regular car batteries don’t have internal fuses so our batteries don’t need them. Could they be thinking fuses are done on the customer side and in hindsight they now realize that’s incorrect.

I have no idea, but when I look at those batteries, I get the feeling that whoever are behind, don't really know how batteries are "typically made", and that they have just grabbed a bunch of off the shelf materials the same way a DIYer would do, rather than getting material-efficient, purpose made components. Because there's no lack in raw materials, it's just that those materials are not used in a proper way to create a reliable product.

Which makes me think that maybe no conscious thought went into the safety of the thing. But even if they did look at car batteries, lead acids and lithium batteries are not really comparable. Lead acid batteries are not as flammable if they are shorted and therefore may not need the same kind of protection. Not saying that a shorted lead acid battery isn't dangerous, because there certainly are high currents, heat and the possibility of hydrogen gasses. But it is more likely to just heat up to a point where the electrolyte begins to boil, and the steam is vented out of the battery, until the battery's ESR increases, and the current drops as a result. There's not much flammable in the cells themselves, since the anode and cathode are mostly lead, and the electrolyte is sulfuric acid in a water solution. And while those fumes are probably not good to breathe, or could generate hydrogen gasses, it's still not as likely to cause a violent fire as a lithium battery where the material in the cells fuels the fire. So there is a good reason that lithium need BMSes and protection circuits, where lead acids are more forgiving to abuse

[Kleinstein]

A rather non reliable protection is not good, especially as the sold the batteries as robust and safe. With those marketing claims they should really have a proper fuse.

I don't know how proper batteries hand this, but ideally I would want temperature sensors at the terminals and the internal BMS to check those temperatures. This would not be that much extra effort and provide some protection for loose terminals. Some batteries already have multiple temperature sensors for the cells.

The other problem with the construction is that it does not look reliable for normal use. This would especially be if the batteries get hot from time to time (e.g. in a RV or with high charging currents) so that plastic creep becomes an issue. The aluminum screws can also have more oxide layer to start with and this way may have enough contact resistance to cause overheating issues with high current, even when new.

[madires]

I have no idea, but when I look at those batteries, I get the feeling that whoever are behind, don't really know how batteries are "typically made", and that they have just grabbed a bunch of off the shelf materials the same way a DIYer would do, rather than getting material-efficient, purpose made components. Because there's no lack in raw materials, it's just that those materials are not used in a proper way to create a reliable product.

I agree. That's a the most likely explanation for the current mess.

[EEVblog]

New video!

[Kleinstein]

After knowing that the batteries don't work well with high current, it is a bit stupid to test the capacitor with a current close to the rated current. It is normal that the batteries would be hot in the test.
A lesson learned from the tests so far is that the batteries don't like high temperature / high current. They may still work fine with lower current.

It would have been smarter to first test them at a lower current, to see the actual capacity.
With the the high current tests one more gets the time when the battery got so hot that extra voltage drop or the BMS temperature terminated the discharge.

[Freesurfer]

I agree. If he wants to do a capacity test, the test should be set up to isolate for that. By running high currents beyond what the batteries are proven to handle, he's not just testing capacity at that point, but a whole chain of potential issues. And when you consider that some of the individual cells seem to be dead, it means the remaining good cells are pushed beyond what they were intended to do and aren't subjected to a fair capacity test. A 0.1C test would have been more appropriate. I think at this point, Will's just trying to expose BB as much as he can with frequent content about the issues.

The one thing that surprised me was that there were so many bad cells when he took the cell packs apart, even in the batteries that supposedly haven't been exposed to very high currents.
They don't have battleborn batteries in Europe afaik, but considering these had a great reputation in the US among campers, they must initially have done something right? Or did everyone just buy into a hype without proven results?

[PlainName]

Or did everyone just buy into a hype without proven results?

If you think the battery is OK then how and why would you test it? When my car battery needs replacing I search for good reviews and then buy a likely good one and fit it. It starts the car, keeps doing that, but I suspect it's not at all stressed. Why would I suspect it? If something did go wrong and I thought I had a good battery, because of the hype, it might take me a while to get around to thinking it could be dodgy after I've checked other potential issues.

So, yes, I think many users buy into the hype, don't have a problem because they're not stressing it, and then reconfirm the hype of how brilliant it is. The odd ones saying the thing is shit are clearly abusing it.

[EEVblog]

After knowing that the batteries don't work well with high current, it is a bit stupid to test the capacitor with a current close to the rated current. It is normal that the batteries would be hot in the test.
A lesson learned from the tests so far is that the batteries don't like high temperature / high current. They may still work fine with lower current.

It would have been smarter to first test them at a lower current, to see the actual capacity.
With the the high current tests one more gets the time when the battery got so hot that extra voltage drop or the BMS temperature terminated the discharge.

Yes, agreed.
The capacity and the high current overheating issues are two different things.
If you are serious about capacity then you get a bunch of characteristic curves at different test currents. At least two, say 10% and 50% of rated current, then another one at say 90% or 100% rated current to stress it.

[Siwastaja]

The capacity and the high current overheating issues are two different things.

Probably, but there's also the possibility that it has seen history of continuous overheating, and the cells closest to the hot terminals have lost capacity due to that damage.

For capacity loss, there are two variants, either actual capacity loss, or increased resistance which makes it appear to have less capacity. But the results on the video show that capacity loss seems dramatic even at fairly low currents, which suggests it's not (only) resistance increase, but actual capacity loss. Either just poor quality cells aging on their own (cycling / calendar), or from heat damage, or both.

[Kleinstein]

The capacity loss can indeed have 2 forms:
One is triggering the cell internal protection, so that some cells no longer contribute at all. The other is the cells overall getting old and tired and a more gradual loss of capacitor over time, e.g. from oxigen diffusing to the cells.

The video showed that quite some cells are dead / high impedance and thus the first mode of capacity loss.

The test with high current can trigger the cell internal protection. Though most of the time used with low current, the one cycle with high current may well have damaged the battery and cause much of the capacity loss. The batteries may well have had Ok capacity before the high current test.
I would expect high current the main reason to make cells fail.

The conclusion is still right: check the capacity to test if the battery is still OK. If the capacity is low it would be no longer safe to use with high current, as the current would be concentrated to fewer cells and the temperature sensor may be on a dead cell.
The other more general conclusion is not to use those batteries with high current over a longer time. The 100 A rating is more short time - maybe better stay below 20 A most of the time.

[Freesurfer]

I did some digging to see if Battleborn batteries are available in Europe. I figured that "Battleborn" isn't really a brand name that would necessarily stick well with Europeans. We don't really buy into badass names or military-themed products, and there isn't as much of a market for "doomsday prepper"-marketed stuff. We prefer cleaner, more neutral names. "Battleborn" wouldn't give European buyers any confidence in the products.

Turns out the same batteries are sold under a different name "Transporter Energy", mainly in the UK. Which is also a more appropriate brand name for the European market I would say.
They have a black case and green lid, but otherwise the lid and terminals seem to have the same shape. (Unless LiFePo4's have standardised form factors)

https://nohma.com/product/battle-born-batteries-uk-transporter-energy-100ah-lithium-leisure-battery/
quote from the linked web page:

Battle Born Batteries UK – AKA Transporter Energy – 100Ah 12V Lithium Leisure Battery
Transporter Energy are Battle Born Batteries UK – they supply the leading Battle Born Batteries from the USA produced in the USA to the UK market. The only change is the name and casing – the technology and high quality remain the same.

There is at least one ebay listing from a German seller.
https://www.ebay.com/itm/125895708285

Would be interesting to see if there are any reports of issues with these. Or if they are tested with other third parties for approval in the European market, which could have picked up on the issues seen in the American battleborns.

[EEVblog]

The capacity and the high current overheating issues are two different things.

Probably, but there's also the possibility that it has seen history of continuous overheating, and the cells closest to the hot terminals have lost capacity due to that damage.

Most definitely.
I meant that they should be treated as different test and characteriaation issues, but can certainly be causal related.

[tooki]

I did some digging to see if Battleborn batteries are available in Europe. I figured that "Battleborn" isn't really a brand name that would necessarily stick well with Europeans. We don't really buy into badass names or military-themed products, and there isn't as much of a market for "doomsday prepper"-marketed stuff. We prefer cleaner, more neutral names. "Battleborn" wouldn't give European buyers any confidence in the products.

FYI, the company name comes from the fact they’re located in the state of Nevada, whose nickname is the “Battle Born State” due to it becoming a state during the Civil War. The Nevada state flag says “Battle Born” on it.

Lots of brands in USA use state-related branding, as state patriotism (as opposed to national patriotism) has increased over the last few decades. Often, the nicknames aren’t really well recognized outside of the state, so they end up being “local brand” signals to state residents and just random names to outsiders.

Also, I think the popularity of “Swiss Army”-branded stuff is a strong counterargument for your claims that Europeans dislike military-themed products and prefer neutral names. Less popular than in USA? Yeah, I’d agree.

[mtwieg]

Yeah while Will's videos are entertaining, his approaches to testing and analysis often annoy me.

For example he did another video tearing down one of those bad batteries:


At 1:50, he shows the measured capacity and ESR of a set of cells and says that 73% of the cells are at 0V (internally fuse popped, I suppose) and the remainder only have half their capacity remaining. Well, that means the battery these cells came from should only be capable of delivering 13.5% of its rated capacity. But in his previous video he stated that the worst of all those batteries delivered 26.8Ahr our of its 100Ahr rating. Obviously he made a mistake somewhere, or just didn't explain things correctly.

edit: oh and I have to bring up this other video where he compares crimped vs soldered lugs:


He compares resistance of cable segments with crimped and soldered lugs, stating that lower resistance is better.
He expresses shock that the soldered cables have much lower resistance than he expected, with one of them having lower resistance than the crimped cable.
Then he tests each cable by intentionally giving it a poor connection at both ends, causing the terminals to heat excessively, and he yanks one the cables in that state. To nobody's surprise, the soldered cable comes apart, the crimped one holds together.
He concludes that crimps are obviously better, ignoring his own resistance test results. No suggestion of adding strain relief to address the possibility of disconnects with soldered cables.
I actually agree with his overall conclusion here, but his reasoning is just really lazy and he has clear biases.

[thm_w]

He expresses shock that the soldered cables have much lower resistance than he expected, with one of them having lower resistance than the crimped cable.
Then he tests each cable by intentionally giving it a poor connection at both ends, causing the terminals to heat excessively, and he yanks one the cables in that state. To nobody's surprise, the soldered cable comes apart, the crimped one holds together.
He concludes that crimps are obviously better, ignoring his own resistance test results. No suggestion of adding strain relief to address the possibility of disconnects with soldered cables.
I actually agree with his overall conclusion here, but his reasoning is just really lazy and he has clear biases.

If you have a soldered connection that overheats, strain relief is not going to solve the issue. The solder will continue to heat and oxidize or flow out of the joint, resulting in a poor connection. Even if it is being held in place with a strain relief. Saw one a few days ago on an overheating xenon bulb.

It is a rare situation but for safety critical applications its clear why you'd want to crimp the wire. Soldered was what 5% better? Its not worth risking for such a small benefit.

[jheissjr]

The soldered terminal in the test was not fully soldered. The only way a 10mm diameter cable is tinned is by dipping it in a solder pot to get full penetration into the cable interior. Same with tinning a bare copper lug. The cable and terminal can be soldered together after they are tinned.

https://youtu.be/4CgJ4Y1FGzs?si=Mit_BhHZUYbwp431&t=129

[uer166]

There do exist large terminals designed for soldering, even in vibration environments like automotive.

You need to make sure the intermetallics aren't brittle, and there is strain relief *past* the point where any solder wicking is possible (sometimes you can use a heatsink designed for the wire to control the wicking).

All in all it's less reliable than an equivalent crimp connection when compared directly, but when done right it's not bad.

[tooki]

The soldered terminal in the test was not fully soldered. The only way a 10cm diameter cable is tinned is by dipping it in a solder pot to get full penetration into the cable interior. Same with tinning a bare copper lug. The cable and terminal can be soldered together after they are tinned.

That’s all true, but doesn’t change anything in the end, since the heat wasn’t generated in the solder joint, but in the loose lug bolt connection. Had it been soldered better, it would just have taken a few seconds longer for the additional solder to melt.

[mtwieg]

I don't disagree with any of the above comments. Again, not saying that Will's conclusion regarding crimps vs solder is wrong, just that his reasoning and explanation is poor (for the soldered cable, did he use a terminal designed for soldering? What sort of solder was used? What soldering process?). So he's not convincing.

[thm_w]

I don't disagree with any of the above comments. Again, not saying that Will's conclusion regarding crimps vs solder is wrong, just that his reasoning and explanation is poor (for the soldered cable, did he use a terminal designed for soldering? What sort of solder was used? What soldering process?). So he's not convincing.

He has no formal EE education so it does make sense that he's lacking on many of the fine details.

I don't know if they make solderable versions of this style of high current connector. I believe the point is more so that people would buy the crimp version, try to save money by not getting a crimper, and soldering them.

The type of solder used shouldn't be relevant.

[tooki]

I don't disagree with any of the above comments. Again, not saying that Will's conclusion regarding crimps vs solder is wrong, just that his reasoning and explanation is poor (for the soldered cable, did he use a terminal designed for soldering? What sort of solder was used? What soldering process?). So he's not convincing.

Again, what is the relevance of any of those questions? The heat wasn’t generated in the solder joint, so the overheating would happen no matter what, in those circumstances.