Let me finish beating this subject to death. In the data Mizuno provided yesterday, you see that after 1.5 hours the system reaches the terminal temperature of 0.5 deg C above ambient. It can never climb higher than that, because losses equal the 0.2 W input from the pump. In other words, at this low power level, the system is functioning as an isoperibolic calorimeter, not adiabatic.
In fact, because the pump is left on all the time, the baseline is always high, so we never measure pump input power. The only thing we use is the increase in temperature *above the baseline*, where the baseline includes the 0.5 deg C from the pump. So this method would never detect pump heat, unless you let the whole system cool overnight with the pump off, and then you turned on the pump and began the test immediately. Mizuno never does that. He always lets it run for while until the reactor surface temperature and water are uniform. - Jed
