Ling, I knew that there was the potential for it to explode, but the timing of that explosion IF it happens, is very hard to pick. The first time it happened to me, i was standing right in front of the fire looking directly at the can, then turned to warm my back and the second i was turned around it exploded. That was years ago and after the night we spent trying to make it explode again and it never happened I forgot about it.

This time, after demonstrating the scientific principle I was trying impart, we tried to see if we could make the effect last longer... and this is the part that I knew would cause the explosion because its basically a build up of pressure vs venting of pressure. If the build up exceedes the venting too greatly, the can explodes. What we didnt bank on was how great the explosion would be, and that depends entirely on a largely variable mix of factors.

All in all, the way I did it first was right, worked properly, and I believe would have continued to work properly too had we not altered the method.