FWIW - When I resoldered the bulb sockets back into place (twice) I was using some Mil-Spec solder I got back when I worked at a private military electronics contracter. This stuff IS "high temp" solder, and I was still able to produce the failure (even under water). I have checked all the connections, none seem to be loose. If I put dummy cells into the holder, I get a complete circuit with the switch on with virtually no resistance. Even while fiddling with everything trying to see if there was a bad connection. I can't seem to find one. I don't see how this could be a batch specific failure.
If the bulb sockets are the same for all of the lights, I could only guess that under the same conditions, the failure will still occur, even though kubolaw says his didn't do it. As StoneDog said, it doesn't take much force for the bulb to fall out. I can't believe the paint on the reflector would get hot enough to bubble and not have the bulb failure we have had here.
The only way to really know what the problem is would be if everyone who has one of these lights and reads this thread tries the "test".
I can't decide whether to send this back to PT so they can see the problem themselves firsthand, or just return it to where I bought it last weekend.
I would like to tell myself this is a batch specific problem, but the engineer in me won't let me fool myself so easily.
StoneDog, let us know what the cust. service dept. says. I have several other PT lights and I think that in general, they make great products that are very durable. At least this is a company that is willing to keep up with the times and bring us new products with the latest, greatest technology (unlike M*gl!ght).