In simulating what's happening to the panel when it's on the roof, what we're going to do is we're going to introduce compressed air through this line here and the compressor outside. It is pumping air into a bladder under here, pushing the glass up. We will determine when the thing has failed either with it cracking, or we will watch as the deflection increases with the pressure increase. So that metre over there will determine pressure, and this one will determine how much the panel has risen. We're going to turn the compressor on, and we're going to start feeding air into the bag, and we will monitor the pressure in the bag on that metre over there, and we will monitor how much deflection we've got in the centre of our panel over there.

We've tested the panel. It came up with 6kPa, the pressures that would've been involved in cyclone Tracy. This panel would have withstood those pressures. So we could turn around and safely say that this panel would have probably stayed on the roof, had the roof stayed on the house. And as you saw from the tests, it went up to six and then the pressure stayed at six. while the deflection kept on going up, which suggested to us something drastic was happening within this structure itself.