20 W is good! That is a lot of heat for most laboratory-scale calorimeters. As long as input is below ~100 W it is hard to imagine they would have difficulty measuring that.
It is actually easier to measure heat from 10 to 100 W than, say, 1000 W. I doubt the s/n ratio improves much at 1000 W. Depending on the size of the device, and the operating temperature, a thing like this might be as significant as Rossi's results. Rossi works on a large scale because he likes to. There is nothing wrong with that. It is better by far than working at a power level below 0.1 W, where things get dicey. Rossi's high power levels impress people. But you understand how science and technology work, you will see that -- a ~20 W stable, controlled reaction at a high temperature using common materials . . . . . . is really as impressive as Rossi's. Okay, that 5 conditions, all of them difficult to meet. Of those 5 stability and control are the most important. Anyone who achieves them, at any power level easy to measure, has made a tremendous breakthrough. Rossi certainly has. Mizuno may have a similar breakthrough. I hope to upload his paper at the time of the conference. I hope it will done to his satisfaction and mine. If not I shall upload the poster, and the paper soon after the conference. - Jed