> I hate to defend the guy when he's set off my fraud detector alarms so much, but thermoelectric converters are what, less than 10% efficient? Wait -- the inp

> I hate to defend the guy when he's set off my fraud detector alarms so much, but thermoelectric converters are what, less than 10% efficient? Wait -- the input power requirement is intended to run a heater, not provide electricity. And I should never have mentioned thermoelectric converters, that only complicates the discussion with no benefit. Because the input power is needed to run a "immersion heater" according to the linked page, and because the output power is heat, this whole account quickly unravels. > He would need some kind of steam-driven turbine to achieve the efficiency needed to provide net power. No problem, there are plenty of those available. But the claims in the "fact sheet" are obviously false. If he really had a working unit that developed a megawatt using dozens of smaller units working in parallel, then why doesn't he just start one small unit with a tiny power kick, then use that unit to start four additional units, four to 16, ad infinitum, like a logical design. But Rossi hasn't bothered to think this deeply. The reason? He's not really going to build it. And the reason for that is it cannot be built.