NI guy = Brian Glass, Applications Engineering. Many thanks to him.
As you see, nominal excess power is 12 to 14 W. Most perturbations are probably caused by changes in ambient temperature this morning when the air conditioning came on. Ambient at present 29.4 deg C. Surface temperature of device is difficult to measure with the IR sensor. I think it is 220 to 240 deg C. They use: A TC at the core A surface mounted TC on tube outer surface Ambient TC The calibration constant is established by turning on the active wire initially and measuring the temperature when it reaches a stable state. In other words, by assuming there is no excess heat at first. The line is very stable and flat so this is a reasonable assumption. Yesterday there were several hours with no excess heat, until they cycled the wire on and off for a while, probably to clean it. They also calibrate with Ar gas for several hours. They do not want to do that too long because the Ar may eventually damage the wire. The second wire is for "indirect heating" during calibration and also during the active run. Not sure what that means. This configuration has not been run in their flow calorimeter because it only works at high temperature and that flow calorimeter does not allow high temperatures. So this has only been detected with thermometry and an Ar calibration. But the effect is quite large and I doubt there is a problem. Even arch-skeptic David Kidwell agrees with me on that. The active constantan wire is ~1.1 m long. It is very thin. The total mass is small, so the power density per cubic centimeter is high. I suppose it is in the same range as Rossi. - Jed
