The heat output measurement process detailed here [0] is deeply fishy. There are simple and accurate ways of measuring output heat, but they were prohibited by the device promoter in favor of a highly unreliable method.

The results are based on using thermal cameras to measure the temperature of the outside of the device, and calculating radiative and convective heat output based on the properties the ceramic material it's made of.

The protocol required by the device promoter was to first measure without fuel at low temperature, then measure with fuel at high temperature. This is fishy. It's possible, for instance, that an impurity in the ceramic causes a thermally insulating film to form at high temperature. There's an obvious way to avoid this error: make a third measurement without fuel at roughly the same high temperature by cranking up the electrical input power. But with such basic checks prohibited, it's impossible to have any confidence in the results.

[0] http://www.sifferkoll.se/sifferkoll/wp-content/uploads/2014/...