[SystemSafety] E-bike battery fires in NY

[M Ellims]

Inserted below...

-----Original Message-----
From: systemsafety [mailto:systemsafety-bounces at lists.techfak.uni-bielefeld.de] On Behalf Of Peter Bernard Ladkin
Sent: 15 November 2022 09:34
To: systemsafety at lists.techfak.uni-bielefeld.de
Subject: Re: [SystemSafety] E-bike battery fires in NY



On 2022-11-15 10:13 , M Ellims wrote:
>> For some perspective - it's not just batteries...

> I would distinguish battery safety from electrical safety in general. The causes of battery fires > are physico-chemical and very different from the causes of electrical-circuitry fires.

The primary cause of most battery fires is not usually the underlying chemistry or physical manufacture, failure rates are quoted between 10E-8 and 10E-11 for individual cells HOWEVER take with a grain of salt as getting good information is difficult!

The primary cause is usually some sort of abuse while in use e.g.
Attempting to charge cells below zero deg. C which can lead to dendrite formation (spikes of lithium metal), vehicles will warm cells up to above zero degrees before allowing charging to start (discharge is OK though).
Other things that can cause cell damage are
- discharge below some specified cell voltage
- overcharging i.e. raising cell voltage about some threshold
- charging or discharging about temperature threshold
- failure to re-balance cells (as they age)
- Using mismatched cells in a battery pack i.e. cells with for example different internal resistances.

Precise nature of issues is somewhat dependant on cell chemistry but the above is a reasonable catalogue.

But a poor pack design coupled with a low cost charger...

>> However the risk of a single cell failing can simply be mitigated in 
>> many cases, research from NASA and NREL suggests that by simply separating cells from
>> each other by 1-2mm almost removes the risk of a chain reaction occurring.

> It depends how big the cell is; it depends how the separation is achieved. Safety also depends upon > the design of the containment structure. When Tesla started out, they claimed their battery design, > following these principles, was such that thermal runaway on a cell was contained. Part of the
> solution to the Boeing 787 problem was a heavy containment structure that more or less undid all of > the claimed weight advantages of lithium-ion main batteries.

Tesla vehicles have actually performed fairly well as far as containment is concerned. It can of course fail when you ram the car into something immovable at high speed.

I think with the 787 is one of the problems is that they didn't allow sufficient space between cells, so there was a clear path for heat transfer between them. It's also possible that the pack designers didn't appreciate how much a cell will expand when it fails. For example a  18650 cylindrical cell will produce something like 25 litters (at STP) of gas if it ignites.

>> It would also probably be better if lithium iron phosphate chemistries 
>> were used

> What are those? Linden's 4th edition (2011) has lithium iron disulphide but
> no lithium iron phosphate amongst the combinations.

Probably the most widely used battery chemistry for electric vehicles (i.e. just about everything built in China).

https://en.wikipedia.org/wiki/Lithium_iron_phosphate

>> nickel-magnesium-cobalt (LFP vs NMC)

> What are those? Again, Linden's 4th doesn't have them.

They and NMA (Nickel-Magnesium-Aluminium) batteries have all but replaced an older generation of cells based on LiCoO2 (Lithium-Cobalt-Oxygen) chemistries as they are far less likely ignite and use much less cobalt.

https://en.wikipedia.org/wiki/Lithium_nickel_manganese_cobalt_oxides#Use_of_NMC_electrodes



PBL

Prof. i.R. Dr. Peter Bernard Ladkin, Bielefeld, Germany
Tel+msg +49 (0)521 880 7319  www.rvs-bi.de