I wrote: They did tell me they test the flow calorimetry before turning on the cell. > That is to say, they make sure the inlet and outlet temperatures are the > same. >
I mean they make sure thermocouples register the same temperature. Of course the temperatures are actually the same, since it is the same flowing water. I meant the experimenter confirms the thermocouples are in agreement, and both are immersed in the flowing water and not wicking heat from the housing. As I mentioned yesterday, if there were constrictions or other sources of heat in the cell large enough to be detected with this system, the outlet temperature would register higher before they turn on the cell. In my experience, ordinary thermocouples are never actually in full agreement. Not to the nearest 0.01°C, anyway. So this means you test them, and you note there is a slight bias one way or the other. You might swap them to see if the bias goes away (bad) or reverses (good!). I would also suggest you immerse both together in ice slurry, and then boiling water, and then in slowly cooling stirred water. There are some things you need to watch out for, but under no circumstances do the thermocouples, flow meter, or watt meter register the wrong answer by a factor or 1000, or 100 or 5 or whatever number is most recently posited by Certain People Who Have Never Done Flow Calorimetry. That's out of the question. When there are severe errors, the instruments typically register nothing -- zip, no connection, or as the thermocouple said during the Three Mile Island disaster: 9997 9998 9999 **** **** **** . . . (Overflow -- they had no idea how hot it was after that, because the computer that did A to D conversion was programmed to report only 4 digits.) Or they register minus 10 million or some other absurd value. It never happens that they go out a factor of 5 and nobody notices. - Jed
