Battery experts were unsurprised by reports Wednesday of a lithium-ion battery fire at a General Motors lab. At the same time, experts reached by Design News cautioned the public not to assume that the new breed of EV batteries is explosive. Warren Mayor Jim Fouts told Reuters that an eight-inch-thick door had been blown out. Five people were injured, and 80 were evacuated from the facility after the fire.
"An incident occurred about 8:45 a.m. Wednesday inside a test chamber at the General Motors Alternative Energy Center during extreme testing of an experimental battery," GM said. "Chemical gases from the battery cells were released and ignited in an enclosed chamber. The battery itself was intact." The automaker emphasized that the batteries are experimental and are not used on the Chevy Volt or any other production vehicle.
It isn't clear whether the doors opened automatically in the presence of a gas release or were blown open by a high-pressure gas buildup. Photos published by WDIV-TV in Detroit show the doors open and unburned. This might indicate they operated as designed.
Design News reached out to knowledgeable engineers to get their take on the incident. Industry experts said it would be uncharacteristic for a lithium-ion battery to explode with such force, even during extreme testing.
"One has to be very careful with the word 'explosion,'" Elton Cairns, battery expert and professor of chemical and biomolecular engineering at the University of California, Berkeley, told us. "There's a popular concept of an explosion, and there's a technical meaning. I seriously doubt that what occurred at GM was an explosion."
@Stuart21 - Your proposal is exactly what is wrong with so much of the world today. I would think that a bit of common sense and root cause analysis would yield a better [for the consumer] solution. I'm not sure how often that Aussies dunk their vehicles, but I will not let that stop me from commenting.
Problem statement: Insurance companies payout too much money to repair water damage to dunked vehicles.
Comment: I hope that you do not have a technical background because the cost to protect the electronics [design, test, implement, increased servicability cost, and upgrade to components in many cases] is not insignificant.
Proposed solution: Pass on to every vehicle purchaser the cost for '24 hour dunk protection'.
1. Make 'vehicle dunking insurance' an option available at extra cost or 'not covered'.
2. Increase the price of your insurance so that it is spread among each of your customers.
3. Sames as #2. Pretend that you are a real insurance company, assess the risk, compute the cost, and set the consumer price.
4. Terminate participation in the sale of automobile insurance.
2007 OKA NEV ZEV was in real life submerged in 3 feet of rainwater and while all other cars parked at that lot in Henderson NV during that flash flood rain storm were total losses (about 18 of them), the car could not only be driven off with still about 18 inches of water in it, it only needed to have the throttle sliding pot 5K resistor replaced as the "dirty water" that got in to it made it not operate "smoothly" there after when it dried out.
The car is 48V DC nominal with 4kW/h battery pack and 12 V DC aux battery for normal car equipment (Both sealed lead acid batteries)
So a well designed EV and actually operate as a submarine completely submerged under any water, the high voltage in other designs will NOT cause much other than some water being electrolized to H2 and O2 but that will just bubble off.
But of course NOT SEALED ECU will most likely fail, the one in OKA is potted in RTV after initial test and "burn-in" and so far only 2 have failed since 2003, both due to customers shorting out the output (450 A fused) while experimenting to make the car go faster than the FMVSS mandated 25 MPH limit for LSV.
More than flood, water or anything else customers EXPERIMENTING with their vehicles are far more common causes of problems - they do it with hopping up Hp on diesels, and "tuning" normal cars to self destruction, yet that seldom makes much news - but lot of Internet noise when the OEM denies warranty repairs on "modified" vehicles.
For starters, whatever it was at GM that went boom was probably not a battery explosion. To get technical about it, in an explosion the propogation is by shock waves, which travel at the speed of sound in whatever media is exploding. And sound travels quite rapidly in a solid. An example of explosions is detonation, known as spark knock, in a gas engine. Normal running is just rapid burning. The blowup could have been caused by hydrogen gas liberated by charging some kind of battery, and being ignited by some accident, such as a small spark. The quite rapid combustion brings about a corresponding increase in temperature and, therfore, pressure, which can do a great deal of damage. Some dynamometer cells are built with "blow-out" doors, so that the rapid combustion resulting from a spray of gasoline igniting will not cause a lot of damage.
The chance of a true explosion in a production type battery is nonexistant because it has been designed out. Things can still burn and get messy, but by the time a product gets near to the prototypem stage the pressure relief system has already been put in place, even if the need for it is only theoretical.
MY bet is that whatever battery caused the incident, it was much earlier in 6the stage, closer to the time when they check to see how many charge-discharge cycles some new battery technology can live through before it loses capacity. And the failure was most likely in the venting system that handles the gasses produced in the charge discharge cycle.
But you would not tell the media that because they would distort it so much that it would lead to a general panic, which, after all, is what the news people love to do.
Good point, Glenn. Yet my kids will probably see some of these wild ideas come true, such as cars that drive themselves and won't crash into each other. Those kinds of advances may be expensive, but in lives saved, it's the equivalent of ending a war.
@Daniel- Agree. Recently a Rolls Royce facility was pretty much demolished when a "blade off" test of a jet engine didn't go as expected. It was a test, and I'm sure the results will contribute to a safer design more than if the test had the expected result.
I have a bit of experience with the structural assembly of cars and trucks.
Waterproofing the computer is only the tip of the iceberg. The electric window motors, the radio, the navigation system, the upholstery, the power seat motors, the heater / air conditioner fan and vacuum controls, the electric fan(s) for the radiator, the remote trunk open, the a/c compressor clutch, the power sunroof, the power door locks, the air bags and sensors. The list could go on and on.
Do you propose to include all of these ? Otherwise what you have is a car that has a computer that can be salvaged, while the rest is a loss.
Yep got to love the Hype.... The problem as many of you mentioned is our media. Back in the day it was all about the truth and having the story correct... Now it's all about the best story (regardless of the facts behind it, if it sounds good put it on the front line). Now To be honest it is sort of hilarious. I'm a test engineer and I do sit down and asses the risk that each test has. Overheating a battery in a closed unvented room might have raised a few flags on my review... Just Saying..
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.