I've only been a member of this forum for a relatively short while, but I've read so many sad crash stories. It is very surprising to me the number of pilots who report that their batteries were found next to their crashed birds. Or that the battery was found to be partially ejected from the aircraft. Take a look at the latching mechanisms on the battery. There's no taper to them. When that battery is snapped into place those two laches have a death grip on the housing. That is in addition to the somewhat tight fit of the battery to the housing even without the latches. Now try pulling your battery out without depressing the latches. I've got a fairly healthy grip and can easily pull with at least 30 lbs of force. I could pull harder but don't want to risk damaging the frame. I can't budge that battery. So how is it that they are 'falling out' on impact? Assume for a minute that the bird falls straight down onto concrete and that it lands directly on the side where the battery ejects. The first thing that hits is two props, and then the two arms. How many G's are experienced by the battery after this somewhat shock absorbed worst case scenario? Ten? Fifteen, twenty? That means that up to 20 pounds of battery weight is pushing against those latches. I can't budge the battery with thirty. How can 20, in our worst case scenario, pop the battery out? My theory is that pilots are not ensuring that their batteries are snapped into place and that they are vibrating free. The tight fit? Easily overcome by a process similar to liquefaction of the ground during an earthquake. When high frequency vibration is induced into a surface it tends to greatly reduce the friction between any two components in contact with each other. Just as we saw in so many cases of runaway acceleration of cars, the drivers swear that their foot was slammed down on the brake when it was on the accelerator, I think pilots staring at the remains of their expensive birds are loathe to think operator error. It's human nature. Many of those crashed birds, I don't say all, lost power when the battery contacts finally lost...contact. No power...end of flight. Has anyone done a failure analysis taking a look at the data recorders of crashed birds for voltage just as the birds became unresponsive? It would have to carefully ensure at what point voltage went to zero. Did it go to zero at the precise instant of ground contact, or did it some seconds before? If it happened seconds before, that would seem to support my hypothesis.