David Roberson <dlrober...@aol.com> wrote:
> I have a difficult time accepting the premise that the power is constantly > being generated during the complete period from this figure. It is much > more likely to be restricted to .3 hours maximum. Have you given this > figure much thought? > You can see for yourself it is constantly being generated. There is no doubt about it. If there were no sources of heat in the cell, after the pulse the temperature would fall right back to where it was before the pulse. You can compute where it would hit the line, using Newton's law of cooling. It does not do that. It continues to rise all day, until evening when the heat peters out and it begins to fall. (It would not return to ambient, because the pump heat holds it 0.6 deg C above that.) > > I do not expect the anonymous heat to be proportional to input power in > any linear fashion. Also, the time domain emission of that heat will not > match the input. My model does not really care about the exact shape of > the input pulse at this point, only the number of joules emitted. > > Thanks for smoothing out the data for me. What I see looks fairly clean. > > > I realize that there remains a major difference in opinion between you and > Gigi concerning the pump heating. I want to remain out of that argument > but need the best proven information to use for my model. He has done > extensive curve fitting and I have asked him to prove his case better. > He can do curve fitting until the cows come home! Mizuno measured the pump heat in an actual test. I uploaded his data and graph. The temperature does not rise after 1.4 hours. It does not rise in the last 3 hours when ambient is stable. If ambient remained stable, it would never rise or fall, period. Any source of heat will always reach a terminal temperature where the heat leaks equal the heat generation. The test proves Mizuno is right. All the curve-fitting and blather in the world cannot disprove what you see in that graph. Gigi is confused because he does not understand the difference between a decline in ambient and an increase in cell temperature. > Jed, the system time constant is a bit less than 6 hours. That means that > it takes several of these periods before an input no longer effects the > final temperature. You mean for it to return back down to ambient + 0.6 deg C. Yeah, but with this method, energy is measured by a *rising* temperature, not a stable temperature. This is not an isoperibolic calorimeter. Given enough time it would rise until it reaches the peak, but we never give it enough time with this test. Instead, the heat peters out and the temperature begins to fall. It would take several hours -- all night in fact -- for it to cool down back to ambient, but once the anomalous heats cuts out, you can clearly see that has happened because the temperature stops climbing and starts to fall. The previous tests were in isoperibolic mode lasting days or weeks, with less insulation. That was a whole different kettle of fish. - Jed
