I hope you do not mind "reprints", for this is based on my comments from a thread on the same topic in another forum:
From the Interim Report:
The APU controller (discussed in section 1.6.5) monitors the parameters that are needed to operate the APU. The APU controller is powered by the APU battery bus, which receives its power from the APU battery. If the APU battery fails, then the APU battery bus will no longer receive power, and the APU will shut down.
And the ventilation fans which will remove any smoke resulting from the event will shut down too.
The ventilation fans in question are probably powered by multiple sources, including the total of 4 engine driven generators, and during flight the fans would therefore likely work OK.
The problem here is definitely that when the battery is both the only power source for the fans and the battery is also the source of the smoke (or perhaps there is NO battery power for the fans) you have an unsurmountable logical problem.
It is much less likely that the ventilation would have failed had the plane been in flight, especially as there would, at least in the new design, have been some ventilation from cabin pressurization and low pressure on the outside of the flame thrower -- I mean vent duct.
This seems even worse design to me though:
Firefighters reported that removing the battery was difficult because a metal kick shield installed in front of the battery prevented them from accessing the battery’s quarter-turn quick disconnect knob. Also, the quick disconnect knob could not be turned because it was charred and melted.
!!!:
The stainless steel sleeve and signal wires had damage consistent with excessive electrical current where they attached to the connectors at each end; at the battery case, the damage was also consistent with fire.
Excessive current on the signaling wires to the battery charger? Maybe just ground current in the signal grounds and the shield because of internal short circuit inside the battery case? At any rate, it does not appear to have damaged the charger.
Only two temperature sensors in the box (thermistors), possibly one for each of two BMU card? The 8 wires from each BMU card were described as voltage sensing wires, but probably doubled as balancing current conductors. The two cards did not contain two independent Battery Management Systems (BMSs) but were rather parts of one unit Battery Management System.
FDR data showed that, at 1021:01, a 1-volt decrease from the designed voltage of the APU battery (32 volts) was recorded. Three seconds later, the data showed a change
in current flow to 44 to 45 amperes into the battery. The battery voltage continued to decrease, and, at 1021:08, the current flow returned to 3 amperes out of the battery. At 1021:30, the battery voltage decreased to 28 volts, and the APU shut down 7 seconds later.30 Table 2 shows selected events recorded before and after the APU shutdown. The FDR did not record any data indicating that the APU battery voltage had exceeded 32 volts.
Not that it necessarily contributed to the event, but the initial sudden 1 volt anomaly did not cause the charger to shut down! This continued for 29 seconds. The steady decrease in battery voltage while charging also was not seen as significant!! The charge rate indicates that for whatever reason (probably the routine APU startup 15 minutes earlier) the battery was still substantially discharged at the time of the event. It was also being rapidly recharged after the substantial load of the APU start. No temperature information seems to have been recorded (?). I am also open to speculation that if the original voltage drop represented an internal short circuit in one cell, then the remaining cells could easily have been overcharged without the charger realizing it from the series voltage. Have to look at the BMU data to be sure.
OOPS:
The BMU main circuit card and sub-circuit card do not contain nonvolatile memory (NVM), and none of the BMU data are recorded on the FDR.
I think that the current flow reported is from the hall effect sensor, positioned like a shunt, in the battery itself and so represents net charging current to the battery independent of what the charge controller may have been delivering to the bus at the same time.
Aft EAFR stopped recording. Forward EAFR continued recording for about 9 minutes
58 seconds.
When the battery went offline and the APU spun down, there was no longer a source of power to the rear EAFR.
And, the last of my stream of consciousness commentary for now:
The battery shall be designed to prevent spilling flammable fluid, a hazardous event with occurrence with a probability of less than 10-9.
Much has been made in the press with the fact that two battery events have occurred within the first 50,000 flight hours that should have had a probability of 1 in 10,000,000 hours or less. If you think that the evidence of either of the two events supports that flaming electrolyte was in fact discharged, we have now hit the 1 in 1,000,000,000 condition too! Note that it was not just flammable but actually burning fluid. Maybe that's safer?