Any data or estimates as to the volume inside the Rossi device, available to be filled with water up to the exit hole, and the additional space above the maximum water level, available to be filled up with mist, foam, froth, bubbles, and steam?
If the available water volume is, say, 180 cc, then a flow of 2 cc/sec would take 90 seconds to fill it and then start filling the 9 m black output hose. I wonder if some of the kinks in the water temperature measured by the thermister (which is only one data location in a complex witch's cauldron) might turn out to reflect a possibly intricate water flow. Could steam vapor blocks form to block and switch some of the flows? Data! More data! Give me data, I say!
