In the no, really? file today we have this:
Short-circuits inside a battery triggered a fire aboard a parked Boeing Co. Dreamliner last month, said U.S. investigators, who also found the safety approval process for the power devices was flawed and "must be reconsidered" before the jets return to passenger service.
This is what I said originally:
The fact that overcharging is not implicated in at least one of these incidents leads to some uncomfortable places, with the most-serious being a potential problem with the assembly of the battery itself (that is, an internal flaw that produced a short circuit inside the case.) This has occurred with laptop and similar batteries in the past and is extremely dangerous because the usual protective circuits are of course not effective. If this is proved to be the case here then Boeing's supplier of that battery likely has a major problem with future orders, but more importantly Boeing may have an unsolveable problem with the use of these batteries, because it is essentially impossible to insure that no defects in the battery pack's manufacturing will ever occur.
And now it appears that's exactly what happened.
The next question is how the battery design was approved in the first place. That trail leads to some uncomfortable places as well. See, Boeing (like all other makers of such things) is generally permitted to "self-certify" a lot of safety-critical design parameters, performing their own alleged testing and submitting their claims to the government.
These entire purpose of using lithium battery chemistry in the first place is to gain energy density -- that is, the amount of power that can be stored for a given mass and size.
But isolation of the cells via containers and physical space to prevent propagation of a failure between cells reduces that energy density! Therefore you have a dance of sorts; your goal is maximum energy density or you'd use a chemistry that didn't have the same sort of risks that these cells do. Yet to provide physical isolation you need both space and mass, which goes directly against your "prime directive" in using these things in the first place.
Boeing has been cleared to start closely-monitored test flights to try to identify exactly why these failures are happening. But there are a lot of people who still seem to think that there's an answer that can be found with the existing design and without any sort of material modification.
My problem with this process is that in a car or other land-based conveyance you can get out and run away if there's a fire, and the risk of such an event is directly related to the energy density in the device and how fast it can be delivered. The entire point of using these batteries is to pack a lot of energy that can be delivered quickly (to start the APU, for example) in a small space and mass.
While the risk of one of these planes being downed by a battery failure is relatively small, the fact remains that lithium battery fires are hellishly difficult to put out, and since this is considered a critical component that cannot fail in flight such a failure, even if you CAN prevent a fire, is considered unacceptable.
So how do we get from where we are now to where we need to be given the existing design?
I'm skeptical that you can, and given the record of our government when it comes to utterly outrageous and unsupportable acts that benefit themselves and their "friends" and screw or worse -- kill citizens-- I am by no means comforted by this alleged "review process" nor am I willing to accept the FAA's alleged "all clear" on this issue, if and when it comes, without strict proof.