I think the Boeing fix(es) are the wrong approach and are risky.
Reason why: Saw a documentary the other day on a fault on Boeing777 plane, that resulted in a British Airways plane landing short of the runway (crash landing). Pilots managed to land it without any fatalities, but it was a very near miss. The aircraft crash investigators thought they would have the answers quickly, but it actually took them over 1 year to REPRODUCE THE FAULT and implement the RIGHT fix. When the autopilot landing the 777 requested more power, both engines responded with less power! That caused the incident. After much work, the investigators concluded it must be ice clogging the fuel lines that starved the engines of fuel. When they recreated the temperatures experienced by that particular flight, they found at vey cold temperatures (-30C) hardly any ice formed. At higher temperature (-20C) then much more ice formed and it stuck to the fuel line pipes. Still, this did not create the problem. ONLY when a sudden increase in fuel demanded was issued, did the ice get knocked off the pipe walls and land onto the intake of the heating system, that is designed to heat the fuel before sending it to the engines. This heater had a number of small pipes sticking up at varying heights and the ice got stuck there! Re-designing to make the pipes all stick up the same amount solve the problem.
The issue is this: no engineer looking at those pipes suspected a problem. It was only REPRODUCING THE FAULT in their test arena did they understand the ONE, simple fix that was actually needed. Imagine if they had taken the Boeing approach and just implemented a scatter-gun of "fixes" - would they have included the one crucial fix? I think not.
I'm not sure that I understand why it should be surprising that managers and PR folk speak in non-engineering terms? How is this surprising? If the press get their stories from management and PR people, that's the lingo they will hear - every time.
If the new packaging has reduced the risk of smoke or fire to something lower than that associated with other batteries in use, then surely that has to be acceptable. Even if people aren't conscious of it, they take risks from the time they are conceived.
It would be interesting to know if 1) batteries were produced in U.S. or another country. This may speak as to quality not only of test batteries , but also Production 2) If altitude/cabin pressurization testing (at least in a pressuritzation chamber) was done.
I have not read a lot on the details of the failure and the testing methodology, but many folks have said that these batteries were prone to problems. It would be interesting to know how much QC and Testing was done to try and mitigate the final product problems.
Blog Doing Math in FPGAs Tom Burke 21 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...