From: bill.sloman@ieee.org   
      
   On 21/12/2025 1:39 am, Liz Tuddenham wrote:   
   > Bill Sloman  wrote:   
   >   
   >> On 20/12/2025 10:02 pm, Liz Tuddenham wrote:   
   >>> Bill Sloman  wrote:   
   >>>   
   >>>> On 19/12/2025 6:49 am, Liz Tuddenham wrote:   
   >>>   
   >>> [...]   
   >>>>> Warning the user isn't much good, the battery technology needs to be   
   >>>>> fail-safe not impending-fail-evident to the user.   
   >>>>   
   >>>> Fail safe would involve a big resistor into which you could start   
   >>>> discharging the battery if you detected worrying warming. You'd have to   
   >>>> design the system to cope with that, and it would  make the designers   
   >>>> job more difficult.   
   >>>   
   >>> Let's do some sums:   
   >>   
   >> First show where you got your numbers from.   
   >>   
   >> I've snipped out that bit of bizarre speculation.   
   >   
   > In more detail: the delamination of the seperator occurs at 25 metres   
   > per second but the thermal runaway reaches a peak of 600 mm/sec and then   
   > falls to 80 mm/sec according to Franson, Pfaff et al. "Exploring thermal   
   > runaway propagation in Li-ion batteries through high-speed X-ray imaging   
   > and thermal analysis".   
   >   
   > For their experiment, they initiated the failure by penetration with a   
   > nail, but the same propagation could equally well be started by failure   
   > of a very small area of a separator.  The nail penetration was near the   
   > casing and this sometimes resulted in a hole melting in the casing and   
   > relieving the excess internal pressure.  A separator failure  away from   
   > the casing could well result in much higher pressures and greater   
   > spreading of incandescent materials   
   This wasn't just an experiment, but rather a piece of fossil carbon   
   industry propaganda generated by simulating a remarkably improbable   
   incident.   
   > They measured the propagation time between the initially-failed cell and   
   > an adjacent cell to be about 4 minutes but various videos of lithium   
   > battery fires show cells exploding at a faster rate than this, once the   
   > fire has taken hold.   
   >   
   > If we take the 4-minute figure as a reasonable approximation, this is   
   > the time in which a 70 kWh battery must be discharged to prevent a   
   > failed cell from setting off the others.  That is more than 1 megawatt   
   > to be dissipated in something the size of a car.   
      
   This is roughly equivalent to insisting that petrol-engined car won't   
   catch fire if you shoot a bullet into the petrol tank.   
      
   >> In reality, the problem is picking up the increased rate of   
   >> self-discharge long before you get to the point where thermal runaway is   
   >> likely - the battery has to get above 120C before this can get going.   
   >   
   > A typical cell holds around 80 Wh of energy but less than 1 watt could   
   > easily heat a small area of separator to over 120C without the   
   > temperature rise or the discharge current being detectable outside the   
   > cell.   
      
   That depends on your temperature sensing. Any useful battery condition   
   monitor has have at least one temperature sensor inside the battery, and   
   another one for external temperature. 120C is a lot hotter than ambient   
   anywhere you are going to use a car, and the battery as a whole would   
   start warming up fast pretty much instantly. Getting millidegree   
   sensitivity out of a temperature sensor is trivial.   
      
   >  if you think you know a way of reliably detecting the failure of   
   > less than a square millimetre of separator in a battery containing 500g   
   > of materials, including about half a square metre of separator, the car   
   > industry would be glad to hear from you.   
      
   They wouldn't. Anybody who understands the physics involved could tell them.   
      
   > If you don't know of such a system, your assertions that lithium   
   > batteries are safe as long as the designer has done his (or her) job   
   > properly, and they can be discharged before a failure become   
   > catastrophic, are based on nothing more than wishful thinking.   
      
   Not really. The grand-daughter of one of my old friends married a guy   
   who'd done his Ph.D. on lithium batteries and is now a lecturer at the   
   University of New South Wales, and I got to talk to him about the   
   subject from time to time. The marriage hasn't lasted and the   
   grand-daughter - who works for Microsoft like two of her uncles and a   
   couple of her cousins - is moving the Microsoft headquarters in   
   Washington state, so I won't get any extra information form there.   
      
   >>> The battery capacity of cars, on average, is about 70 kWh.  This means a   
   >>> resistor capable of dissipating 70 kW continuously is needed to   
   >>> discharge the battery in one hour.   
   >>   
   >> You'd dump the excess energy slowly into the motor, letting it rock the   
   >> car rapidly back and forth by about a foot or so to generate a little   
   >> extra air circulation. It would take a while to discharge the battery,   
   >> but it would get it done.   
   >>   
   >> It would be a emergency solution - the driver would get told that the   
   >> battery needed attention long before this would be justifiable, and in   
   >> our brave new world the battery condition monitor would probably have   
   >> it's own mobile phone to rat out the inattentive owner to the local fire   
   >> service.   
   >>   
   >>> I'm sure cars with a red-hot bedstead of resistance wire on the roof   
   >>> would soon catch on.   
   >>   
   >> Your enthusiasm for impractical solutions is noted.   
   >   
   > It is probably just as practical as having a car start rocking backwards   
   > and forwards for hours on end to discharge the battery.   
      
   It's perfectly practical as an "if all else fails" solution. It should   
   never be need to be invoked, but there are plenty of silly people around   
   and we do have to cope with their inadequacies.   
      
   My car, which was bought new in 2011, has little radar sensors fore and   
   aft that could tell it how far it could go back and forth without   
   hitting anything.   
      
   > An even better solution (in a Brave New World) would be to have it drive   
   > itself to somewhere where it can't cause any harm, as quickly as   
   > possible.  Perhaps every Local Authority should have a designated place,   
   > downwind of the town, where cars with faulty batteries could be   
   > programmed to drive themselves and burn out in relative safety.   
      
   Cars are frequently garaged in closed spaces. They don't tend to be   
   delivered with the hardware that would let them get themselves out.   
   Mobile phones work inside garages. There are garages and car-ports in   
   Australia that lie beyond the reach of the mobile network, but the fire   
   service out there aren't all that local.   
      
   --   
   Bill Sloman, Sydney   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)